This document proposes and defines the problem of privacy-preserving multi-keyword ranked search over encrypted cloud data (MRSE). It establishes strict privacy requirements for such a system, including data privacy, index privacy, keyword privacy and trapdoor privacy. It presents the MRSE framework with four algorithms: Setup, BuildIndex, Trapdoor and Query. The Query algorithm allows cloud servers to perform a ranked search on encrypted indexes and return similarity-ranked results, while preserving privacy.
Privacy preserving multi-keyword ranked search over encrypted cloud data 2Swathi Rampur
This document proposes and defines the problem of privacy-preserving multi-keyword ranked search over encrypted cloud data (MRSE). It establishes strict privacy requirements for such a system, including data privacy, index privacy, keyword privacy and trapdoor privacy. It presents the MRSE framework with four algorithms: Setup, BuildIndex, Trapdoor and Query. The Query algorithm allows cloud servers to perform a ranked search on encrypted indexes and return similarity-ranked results, while preserving privacy.
Enabling efficient multi keyword ranked search over encrypted mobile cloud da...redpel dot com
This document summarizes a research paper that proposes a new efficient multi-keyword ranked search (EMRS) scheme over encrypted mobile cloud data through blind storage. The key contributions are:
1) It introduces a relevance score to the searchable encryption scheme to enable multi-keyword ranked search over encrypted documents. It also constructs an efficient index to improve search efficiency.
2) It modifies the blind storage system to solve the trapdoor unlinkability problem and conceal the search user's access pattern from the cloud server.
3) It provides security analysis showing the EMRS can achieve confidentiality of documents/index, trapdoor privacy, unlinkability, and conceal access patterns while experiments show it improves efficiency over existing proposals.
This document proposes a system for enabling secure and efficient ranked keyword search over outsourced cloud data. It summarizes that existing searchable encryption techniques only support basic Boolean search and do not consider relevance ranking. The proposed system explores using relevance scores from information retrieval to build a searchable index and develops a one-to-many order-preserving mapping technique to protect sensitive score information, allowing efficient server-side ranking without compromising keyword privacy. The system provides ranked search results while maintaining strong security guarantees.
Survey on Privacy- Preserving Multi keyword Ranked Search over Encrypted Clou...Editor IJMTER
The advent of cloud computing, data owners are motivated to outsource their complex
data management systems from local sites to commercial public cloud for great flexibility and
economic savings. But for protecting data privacy, sensitive data has to be encrypted before
outsourcing.Considering the large number of data users and documents in cloud, it is crucial for
the search service to allow multi-keyword query and provide result similarity ranking to meet the
effective data retrieval need. Related works on searchable encryption focus on single keyword
search or Boolean keyword search, and rarely differentiate the search results. We first propose a
basic MRSE scheme using secure inner product computation, and then significantly improve it to
meet different privacy requirements in two levels of threat models. The Incremental High Utility
Pattern Transaction Frequency Tree (IHUPTF-Tree) is designed according to the transaction
frequency (descending order) of items to obtain a compact tree.
By using high utility pattern the items can be arranged in an efficient manner. Tree structure
is used to sort the items. Thus the items are sorted and frequent pattern is obtained. The frequent
pattern items are retrieved from the database by using hybrid tree (H-Tree) structure. So the
execution time becomes faster. Finally, the frequent pattern item that satisfies the threshold value
is displayed.
Privacy preserving multi-keyword ranked search over encrypted cloud dataIGEEKS TECHNOLOGIES
This document proposes a system called privacy-preserving multi-keyword ranked search over encrypted cloud data (MRSE). Existing searchable encryption systems only support single-keyword or boolean keyword search without result ranking. The proposed MRSE system allows a user to search for multiple keywords and returns documents ranked by relevance. It establishes privacy requirements and uses an efficient "coordinate matching" semantic to quantify document similarity based on keyword matches. The system architecture includes modules for data owners to encrypt and upload files, for users to search and download encrypted files, and for ranking search results.
Fuzzy Keyword Search Over Encrypted Data in Cloud ComputingIJERA Editor
As Cloud Computing becomes prevalent, more and more sensitive information are being centralized into the cloud. For the protection of data privacy, sensitive data usually have to be encrypted before outsourcing, which makes effective data utilization a very challenging task. Although traditional searchable encryption schemes allow a user to securely search over encrypted data through keywords and selectively retrieve files of interest, these techniques support only exact keyword search. This significant drawback makes existing techniques unsuitable in cloud computing as it is greatly affect system usability, rendering user searching experiences very frustrating and system efficiency very low. In this paper, for the first time we formalize and solve the problem of effective fuzzy keyword search over encrypted cloud while maintaining keyword privacy. In our solution, we exploit edit distance to quantify keyword similarity and develop new advanced technique on constructing fuzzy keyword sets which greatly reduces the storage and representation overheads. In this way, we show that our proposed solution is secure and privacy preserving, while realizing the goal of fuzzy keyword search.
IRJET- Privacy Preserving Keyword Search over Cloud DataIRJET Journal
The document proposes a scheme for secure ranked keyword search over encrypted cloud data. It discusses encrypting data before outsourcing it to the cloud for confidentiality. An index of keyword sets is stored on a local trusted server, while encrypted data files are stored on an untrusted cloud server. When a user searches for keywords, relevant files are ranked based on criteria like term frequency and file length. The top results are then retrieved from the cloud without revealing sensitive information to unauthorized parties. The system aims to enable efficient yet private keyword searches on large amounts of outsourced encrypted data.
Privacy preserving multi-keyword ranked search over encrypted cloud data 2Swathi Rampur
This document proposes and defines the problem of privacy-preserving multi-keyword ranked search over encrypted cloud data (MRSE). It establishes strict privacy requirements for such a system, including data privacy, index privacy, keyword privacy and trapdoor privacy. It presents the MRSE framework with four algorithms: Setup, BuildIndex, Trapdoor and Query. The Query algorithm allows cloud servers to perform a ranked search on encrypted indexes and return similarity-ranked results, while preserving privacy.
Enabling efficient multi keyword ranked search over encrypted mobile cloud da...redpel dot com
This document summarizes a research paper that proposes a new efficient multi-keyword ranked search (EMRS) scheme over encrypted mobile cloud data through blind storage. The key contributions are:
1) It introduces a relevance score to the searchable encryption scheme to enable multi-keyword ranked search over encrypted documents. It also constructs an efficient index to improve search efficiency.
2) It modifies the blind storage system to solve the trapdoor unlinkability problem and conceal the search user's access pattern from the cloud server.
3) It provides security analysis showing the EMRS can achieve confidentiality of documents/index, trapdoor privacy, unlinkability, and conceal access patterns while experiments show it improves efficiency over existing proposals.
This document proposes a system for enabling secure and efficient ranked keyword search over outsourced cloud data. It summarizes that existing searchable encryption techniques only support basic Boolean search and do not consider relevance ranking. The proposed system explores using relevance scores from information retrieval to build a searchable index and develops a one-to-many order-preserving mapping technique to protect sensitive score information, allowing efficient server-side ranking without compromising keyword privacy. The system provides ranked search results while maintaining strong security guarantees.
Survey on Privacy- Preserving Multi keyword Ranked Search over Encrypted Clou...Editor IJMTER
The advent of cloud computing, data owners are motivated to outsource their complex
data management systems from local sites to commercial public cloud for great flexibility and
economic savings. But for protecting data privacy, sensitive data has to be encrypted before
outsourcing.Considering the large number of data users and documents in cloud, it is crucial for
the search service to allow multi-keyword query and provide result similarity ranking to meet the
effective data retrieval need. Related works on searchable encryption focus on single keyword
search or Boolean keyword search, and rarely differentiate the search results. We first propose a
basic MRSE scheme using secure inner product computation, and then significantly improve it to
meet different privacy requirements in two levels of threat models. The Incremental High Utility
Pattern Transaction Frequency Tree (IHUPTF-Tree) is designed according to the transaction
frequency (descending order) of items to obtain a compact tree.
By using high utility pattern the items can be arranged in an efficient manner. Tree structure
is used to sort the items. Thus the items are sorted and frequent pattern is obtained. The frequent
pattern items are retrieved from the database by using hybrid tree (H-Tree) structure. So the
execution time becomes faster. Finally, the frequent pattern item that satisfies the threshold value
is displayed.
Privacy preserving multi-keyword ranked search over encrypted cloud dataIGEEKS TECHNOLOGIES
This document proposes a system called privacy-preserving multi-keyword ranked search over encrypted cloud data (MRSE). Existing searchable encryption systems only support single-keyword or boolean keyword search without result ranking. The proposed MRSE system allows a user to search for multiple keywords and returns documents ranked by relevance. It establishes privacy requirements and uses an efficient "coordinate matching" semantic to quantify document similarity based on keyword matches. The system architecture includes modules for data owners to encrypt and upload files, for users to search and download encrypted files, and for ranking search results.
Fuzzy Keyword Search Over Encrypted Data in Cloud ComputingIJERA Editor
As Cloud Computing becomes prevalent, more and more sensitive information are being centralized into the cloud. For the protection of data privacy, sensitive data usually have to be encrypted before outsourcing, which makes effective data utilization a very challenging task. Although traditional searchable encryption schemes allow a user to securely search over encrypted data through keywords and selectively retrieve files of interest, these techniques support only exact keyword search. This significant drawback makes existing techniques unsuitable in cloud computing as it is greatly affect system usability, rendering user searching experiences very frustrating and system efficiency very low. In this paper, for the first time we formalize and solve the problem of effective fuzzy keyword search over encrypted cloud while maintaining keyword privacy. In our solution, we exploit edit distance to quantify keyword similarity and develop new advanced technique on constructing fuzzy keyword sets which greatly reduces the storage and representation overheads. In this way, we show that our proposed solution is secure and privacy preserving, while realizing the goal of fuzzy keyword search.
IRJET- Privacy Preserving Keyword Search over Cloud DataIRJET Journal
The document proposes a scheme for secure ranked keyword search over encrypted cloud data. It discusses encrypting data before outsourcing it to the cloud for confidentiality. An index of keyword sets is stored on a local trusted server, while encrypted data files are stored on an untrusted cloud server. When a user searches for keywords, relevant files are ranked based on criteria like term frequency and file length. The top results are then retrieved from the cloud without revealing sensitive information to unauthorized parties. The system aims to enable efficient yet private keyword searches on large amounts of outsourced encrypted data.
iaetsd Preserving private multi keyword searching with ranking by anonymous i...Iaetsd Iaetsd
This document discusses privacy-preserving multi-keyword ranked search over encrypted cloud data. It proposes assigning anonymous IDs to cloud users to hide their identities from the cloud service provider and better protect sensitive data on the cloud. The system allows data owners to encrypt and outsource data to the cloud for storage. It then builds a searchable index to allow authorized users to search for keywords without learning the content. The cloud server ranks search results based on relevance but hides information about important documents. Previous works focused on search and privacy but revealed user identities. The proposed system addresses this by anonymizing user IDs to maintain privacy under two threat models.
The document proposes a system for multi-keyword ranked search over encrypted cloud data while preserving privacy. It addresses limitations in previous systems that allowed single keyword search or did not consider privacy. The proposed system uses asymmetric key encryption, a block-max index, and dynamic key generation to allow efficient retrieval of relevant encrypted data from the cloud without security breaches. It involves three parts: (1) a server that encrypts and stores data in the cloud and sends decryption keys; (2) a cloud server that handles search requests, ranks results, and responds; and (3) users that request data from the cloud server.
An proficient and Confidentiality-Preserving Multi- Keyword Ranked Search ove...Editor IJCATR
Cloud computing has developed progressively prevalent for data owners to outsource their data to public cloud servers
while consenting data users to reclaim this data. For isolation disquiets, a secure rifle over encrypted cloud data has stirred numerous
research mechanisms underneath the particular owner model. Conversely, most cloud servers in practice do not just assist one owner,
as an alternative, their sustenance gives multiple owners to share the assistances carried by cloud computing. In this proficient and
confidentiality-Preserving Multi-Keyword Ranked Search over Encrypted Cloud Data, new schemes to deal with Privacy preserving
Ranked Multi-keyword Search in a Multi-owner model (PRMSM) has been introduced. To facilitate cloud servers to execute secure
search without knowing the actual data of both keywords and trapdoors, we thoroughly build a novel secure search protocol. To rank
the search results and domain the privacy of relevance scores amongst keywords and files. To thwart the assailants from snooping
secret keys and fantasizing to be legal data users submitting pursuits, a novel dynamic secret key generation protocol and a new data
user authentication protocol is discussed.
A SECURE AND DYNAMIC MULTI-KEYWORD RANKED SEARCH SCHEME OVER ENCRYPTED CLOUD...Nexgen Technology
bulk ieee projects in pondicherry,ieee projects in pondicherry,final year ieee projects in pondicherry
Nexgen Technology Address:
Nexgen Technology
No :66,4th cross,Venkata nagar,
Near SBI ATM,
Puducherry.
Email Id: praveen@nexgenproject.com.
www.nexgenproject.com
Mobile: 9751442511,9791938249
Telephone: 0413-2211159.
NEXGEN TECHNOLOGY as an efficient Software Training Center located at Pondicherry with IT Training on IEEE Projects in Android,IEEE IT B.Tech Student Projects, Android Projects Training with Placements Pondicherry, IEEE projects in pondicherry, final IEEE Projects in Pondicherry , MCA, BTech, BCA Projects in Pondicherry, Bulk IEEE PROJECTS IN Pondicherry.So far we have reached almost all engineering colleges located in Pondicherry and around 90km
Enabling efficient multi keyword ranked search over encrypted mobile cloud da...LeMeniz Infotech
Enabling efficient multi keyword ranked search over encrypted mobile cloud data through blind storage
Do Your Projects With Technology Experts
To Get this projects Call : 9566355386 / 99625 88976
Visit : www.lemenizinfotech.com / www.ieeemaster.com
Mail : projects@lemenizinfotech.com
Privacy preserving multi-keyword ranked search over encrypted cloud dataNexgen Technology
The document proposes a solution called privacy-preserving multi-keyword ranked search over encrypted cloud data (MRSE) that allows searching encrypted data stored in the cloud using multiple keywords. It establishes privacy requirements for such a system and uses an efficient "coordinate matching" similarity measure to determine relevance between search queries and documents. It first presents a basic MRSE scheme based on secure inner product computation, then improves upon it with two schemes that achieve different privacy levels while introducing low overhead on computation and communication as shown by experiments on real-world datasets.
This document proposes a scheme called PRMSM that enables privacy-preserving ranked multi-keyword search on encrypted cloud data from multiple data owners. It constructs a secure search protocol that allows cloud servers to perform searches without knowing the actual data or trapdoors. It also proposes a novel function to preserve the privacy of relevance scores between keywords and files during ranking. The scheme supports dynamic key generation, user authentication, and efficient user revocation to enhance security. Experiments show the efficacy and efficiency of PRMSM.
A secure and dynamic multi keyword ranked search scheme over encrypted cloud ...LeMeniz Infotech
A secure and dynamic multi keyword ranked search scheme over encrypted cloud data
Do Your Projects With Technology Experts
To Get this projects Call : 9566355386 / 99625 88976
Visit : www.lemenizinfotech.com / www.ieeemaster.com
Mail : projects@lemenizinfotech.com
Blog : http://ieeeprojectspondicherry.weebly.com
Blog : http://www.ieeeprojectsinpondicherry.blogspot.in/
Youtube:https://www.youtube.com/watch?v=eesBNUnKvws
A secure and dynamic multi keyword ranked search scheme over encrypted cloud ...Pvrtechnologies Nellore
This document proposes a secure multi-keyword ranked search scheme over encrypted cloud data that allows for dynamic operations like deletion and insertion of documents. It constructs a special tree-based index structure and uses a "Greedy Depth-first Search" algorithm to provide efficient search. The scheme utilizes a secure kNN algorithm to encrypt index and query vectors while still allowing accurate relevance scoring between encrypted vectors. It aims to achieve sub-linear search time and flexible handling of document updates.
A secure and dynamic multi keyword ranked search scheme over encrypted cloud ...CloudTechnologies
We are the company providing Complete Solution for all Academic Final Year/Semester Student Projects. Our projects are suitable for B.E (CSE,IT,ECE,EEE), B.Tech (CSE,IT,ECE,EEE),M.Tech (CSE,IT,ECE,EEE) B.sc (IT & CSE), M.sc (IT & CSE), MCA, and many more..... We are specialized on Java,Dot Net ,PHP & Andirod technologies. Each Project listed comes with the following deliverable: 1. Project Abstract 2. Complete functional code 3. Complete Project report with diagrams 4. Database 5. Screen-shots 6. Video File
SERVICE AT CLOUDTECHNOLOGIES
IEEE, WEB, WINDOWS PROJECTS ON DOT NET, JAVA& ANDROID TECHNOLOGIES,EMBEDDED SYSTEMS,MAT LAB,VLSI DESIGN.
ME, M-TECH PAPER PUBLISHING
COLLEGE TRAINING
Thanks&Regards
cloudtechnologies
# 304, Siri Towers,Behind Prime Hospitals
Maitrivanam, Ameerpet.
Contact:-8121953811,8522991105.040-65511811
cloudtechnologiesprojects@gmail.com
http://cloudstechnologies.in/
1) The document proposes an optimized and secured semantic-based ranking approach for keyword search over encrypted cloud data. It aims to improve search accuracy by considering keyword semantics and different keyword forms.
2) An index is created from unencrypted files containing keyword-file mappings and encrypted relevance scores. Files are encrypted before outsourcing to the cloud.
3) The approach analyzes semantics between keywords, performs stemming, and calculates relevance scores. It encrypts the index and files before outsourcing to the cloud to protect data privacy during searches.
iaetsd Preserving private multi keyword searching with ranking by anonymous i...Iaetsd Iaetsd
This document discusses privacy-preserving multi-keyword ranked search over encrypted cloud data. It proposes assigning anonymous IDs to cloud users to hide their identities from the cloud service provider and better protect sensitive data on the cloud. The system allows data owners to encrypt and outsource data to the cloud for storage. It then builds a searchable index to allow authorized users to search for keywords without learning the content. The cloud server ranks search results based on relevance but hides information about important documents. Previous works focused on search and privacy but revealed user identities. The proposed system addresses this by anonymizing user IDs to maintain privacy under two threat models.
The document proposes a system for multi-keyword ranked search over encrypted cloud data while preserving privacy. It addresses limitations in previous systems that allowed single keyword search or did not consider privacy. The proposed system uses asymmetric key encryption, a block-max index, and dynamic key generation to allow efficient retrieval of relevant encrypted data from the cloud without security breaches. It involves three parts: (1) a server that encrypts and stores data in the cloud and sends decryption keys; (2) a cloud server that handles search requests, ranks results, and responds; and (3) users that request data from the cloud server.
An proficient and Confidentiality-Preserving Multi- Keyword Ranked Search ove...Editor IJCATR
Cloud computing has developed progressively prevalent for data owners to outsource their data to public cloud servers
while consenting data users to reclaim this data. For isolation disquiets, a secure rifle over encrypted cloud data has stirred numerous
research mechanisms underneath the particular owner model. Conversely, most cloud servers in practice do not just assist one owner,
as an alternative, their sustenance gives multiple owners to share the assistances carried by cloud computing. In this proficient and
confidentiality-Preserving Multi-Keyword Ranked Search over Encrypted Cloud Data, new schemes to deal with Privacy preserving
Ranked Multi-keyword Search in a Multi-owner model (PRMSM) has been introduced. To facilitate cloud servers to execute secure
search without knowing the actual data of both keywords and trapdoors, we thoroughly build a novel secure search protocol. To rank
the search results and domain the privacy of relevance scores amongst keywords and files. To thwart the assailants from snooping
secret keys and fantasizing to be legal data users submitting pursuits, a novel dynamic secret key generation protocol and a new data
user authentication protocol is discussed.
A SECURE AND DYNAMIC MULTI-KEYWORD RANKED SEARCH SCHEME OVER ENCRYPTED CLOUD...Nexgen Technology
bulk ieee projects in pondicherry,ieee projects in pondicherry,final year ieee projects in pondicherry
Nexgen Technology Address:
Nexgen Technology
No :66,4th cross,Venkata nagar,
Near SBI ATM,
Puducherry.
Email Id: praveen@nexgenproject.com.
www.nexgenproject.com
Mobile: 9751442511,9791938249
Telephone: 0413-2211159.
NEXGEN TECHNOLOGY as an efficient Software Training Center located at Pondicherry with IT Training on IEEE Projects in Android,IEEE IT B.Tech Student Projects, Android Projects Training with Placements Pondicherry, IEEE projects in pondicherry, final IEEE Projects in Pondicherry , MCA, BTech, BCA Projects in Pondicherry, Bulk IEEE PROJECTS IN Pondicherry.So far we have reached almost all engineering colleges located in Pondicherry and around 90km
Enabling efficient multi keyword ranked search over encrypted mobile cloud da...LeMeniz Infotech
Enabling efficient multi keyword ranked search over encrypted mobile cloud data through blind storage
Do Your Projects With Technology Experts
To Get this projects Call : 9566355386 / 99625 88976
Visit : www.lemenizinfotech.com / www.ieeemaster.com
Mail : projects@lemenizinfotech.com
Privacy preserving multi-keyword ranked search over encrypted cloud dataNexgen Technology
The document proposes a solution called privacy-preserving multi-keyword ranked search over encrypted cloud data (MRSE) that allows searching encrypted data stored in the cloud using multiple keywords. It establishes privacy requirements for such a system and uses an efficient "coordinate matching" similarity measure to determine relevance between search queries and documents. It first presents a basic MRSE scheme based on secure inner product computation, then improves upon it with two schemes that achieve different privacy levels while introducing low overhead on computation and communication as shown by experiments on real-world datasets.
This document proposes a scheme called PRMSM that enables privacy-preserving ranked multi-keyword search on encrypted cloud data from multiple data owners. It constructs a secure search protocol that allows cloud servers to perform searches without knowing the actual data or trapdoors. It also proposes a novel function to preserve the privacy of relevance scores between keywords and files during ranking. The scheme supports dynamic key generation, user authentication, and efficient user revocation to enhance security. Experiments show the efficacy and efficiency of PRMSM.
A secure and dynamic multi keyword ranked search scheme over encrypted cloud ...LeMeniz Infotech
A secure and dynamic multi keyword ranked search scheme over encrypted cloud data
Do Your Projects With Technology Experts
To Get this projects Call : 9566355386 / 99625 88976
Visit : www.lemenizinfotech.com / www.ieeemaster.com
Mail : projects@lemenizinfotech.com
Blog : http://ieeeprojectspondicherry.weebly.com
Blog : http://www.ieeeprojectsinpondicherry.blogspot.in/
Youtube:https://www.youtube.com/watch?v=eesBNUnKvws
A secure and dynamic multi keyword ranked search scheme over encrypted cloud ...Pvrtechnologies Nellore
This document proposes a secure multi-keyword ranked search scheme over encrypted cloud data that allows for dynamic operations like deletion and insertion of documents. It constructs a special tree-based index structure and uses a "Greedy Depth-first Search" algorithm to provide efficient search. The scheme utilizes a secure kNN algorithm to encrypt index and query vectors while still allowing accurate relevance scoring between encrypted vectors. It aims to achieve sub-linear search time and flexible handling of document updates.
A secure and dynamic multi keyword ranked search scheme over encrypted cloud ...CloudTechnologies
We are the company providing Complete Solution for all Academic Final Year/Semester Student Projects. Our projects are suitable for B.E (CSE,IT,ECE,EEE), B.Tech (CSE,IT,ECE,EEE),M.Tech (CSE,IT,ECE,EEE) B.sc (IT & CSE), M.sc (IT & CSE), MCA, and many more..... We are specialized on Java,Dot Net ,PHP & Andirod technologies. Each Project listed comes with the following deliverable: 1. Project Abstract 2. Complete functional code 3. Complete Project report with diagrams 4. Database 5. Screen-shots 6. Video File
SERVICE AT CLOUDTECHNOLOGIES
IEEE, WEB, WINDOWS PROJECTS ON DOT NET, JAVA& ANDROID TECHNOLOGIES,EMBEDDED SYSTEMS,MAT LAB,VLSI DESIGN.
ME, M-TECH PAPER PUBLISHING
COLLEGE TRAINING
Thanks&Regards
cloudtechnologies
# 304, Siri Towers,Behind Prime Hospitals
Maitrivanam, Ameerpet.
Contact:-8121953811,8522991105.040-65511811
cloudtechnologiesprojects@gmail.com
http://cloudstechnologies.in/
1) The document proposes an optimized and secured semantic-based ranking approach for keyword search over encrypted cloud data. It aims to improve search accuracy by considering keyword semantics and different keyword forms.
2) An index is created from unencrypted files containing keyword-file mappings and encrypted relevance scores. Files are encrypted before outsourcing to the cloud.
3) The approach analyzes semantics between keywords, performs stemming, and calculates relevance scores. It encrypts the index and files before outsourcing to the cloud to protect data privacy during searches.
This document summarizes a research paper that assesses collaboration frameworks in multi-cloud environments. It explores the viability of cloud service providers collaborating to offer diverse services without heavy infrastructure spending. The paper reviews several proposed collaboration models and frameworks, including a proxy-based framework using different types of proxies, a unified multi-cloud infrastructure using open service models and configurable federations, and a proxy-as-cloud-broker model using dynamic scheduling algorithms. The paper concludes that multi-cloud environments can eliminate vendor lock-in for consumers and allow services to be accessed based on preference and need rather than a single provider. Key challenges to collaboration include standardization and security across provider platforms and services.
The document proposes a method called RAndom Space Perturbation (RASP) to provide secure and efficient range and k-nearest neighbor (kNN) query services for protected data hosted in the cloud. RASP combines order preserving encryption, dimensionality expansion, random noise injection, and random projection to transform data in a way that preserves the topology of multidimensional ranges, allowing for efficient query processing while providing strong confidentiality guarantees. The authors analyze attacks on the RASP-protected data and queries under a defined threat model and security assumptions. Experimental results demonstrate advantages of the RASP approach in efficiency and security for cloud-based query services.
Este documento presenta el planteamiento del problema de investigación sobre la falta de aplicación de una herramienta para mejorar el desempeño del personal administrativo en la empresa Barroeta Gas en Venezuela. Se argumenta que sin un sistema de mejoramiento laboral, los empleados no tienen claro sus funciones ni la importancia de su trabajo, lo que genera baja productividad. El coaching se propone como una alternativa para motivar al personal a través de una mejor inducción y políticas empresariales. El objetivo general es proponer el coaching como herramienta para mejorar el desempeño del personal
Define and solve the problem of effective and secure ranked keyword search over encrypted cloud data.
Ranked search greatly enhances system usability by returning the matching files in a ranked order regarding to
certain relevance criteria (e.g., keyword frequency), thus making one step closer towards practical deployment of
privecy- preserving data hosting services in Cloud Computing. To improve the security for the data retrieval from
cloud environment, the One Time Password is used. The One Time Passwod is sent to the user mail to view the
original data. The Model exhibits the Querying Process over the cloud computing infrastructure using Secured and
Encrypted Data access and Ranking over the results would benefit the usre for the getting better results.
This document proposes an efficient multi-keyword ranked search (EMRS) scheme over encrypted mobile cloud data through blind storage. The EMRS enables search users to perform multi-keyword searches over encrypted documents stored on a cloud server and receive ranked search results based on relevance. It utilizes techniques like relevance scoring, secure k-nearest neighbor computation, and blind storage to provide search functionality while preserving security and privacy. The scheme is analyzed to demonstrate that it achieves confidentiality of documents and index, trapdoor privacy, trapdoor unlinkability, and conceals access patterns, addressing key security requirements. Experimental results show the EMRS provides improved efficiency and functionality compared to existing proposals.
Efficient Privacy Preserving Clustering Based Multi Keyword Search IRJET Journal
This document proposes an efficient privacy-preserving clustering-based multi-keyword search system. It uses hierarchical clustering to generate clusters of encrypted documents in the cloud. The system aims to improve search efficiency while maintaining security. It utilizes EM clustering, SHA-1 hashing for deduplication, and a user revocation method. Experimental results show the framework has advantages such as efficient memory and time utilization, secure search over encrypted data, secure data storage, and deduplication.
E FFICIENT D ATA R ETRIEVAL F ROM C LOUD S TORAGE U SING D ATA M ININ...IJCI JOURNAL
Cloud computing is an emanating technology allowing
users to perform data processing, use as storage
and data admission services from around the world t
hrough internet. The Cloud service providers charge
depending on the user’s usage. Imposing confidentia
lity and scalability on cloud data increases the
complexity of cloud computing. As sensitive informa
tion is centralized into the cloud, this informatio
n must
be encrypted and uploaded to cloud for the data pri
vacy and efficient data utilization. As the data be
comes
complex and number of users are increasing searchin
g of the files must be allowed through multiple
keyword of the end users interest. The traditional
searchable encryption schemes allows users to searc
h in
the encrypted cloud data through keywords, which su
pport only Boolean search, i.e., whether a keyword
exists in a file or not, without any relevance of d
ata files and the queried keyword. Searching of dat
a in the
cloud using Single keyword ranked search results to
o coarse output and the data privacy is opposed usi
ng
server side ranking based on order-preserving encry
ption (OPE)
Privacy preserving multi-keyword ranked search over encrypted cloud dataShakas Technologies
This document proposes a privacy-preserving approach for multi-keyword ranked search over encrypted cloud data. It establishes strict privacy requirements for such a system that enables encrypted data search services. The approach uses "coordinate matching" semantics to capture document relevance to search queries, and "inner product similarity" to quantitatively evaluate similarity. It presents two schemes for the multi-keyword ranked search that achieve different privacy requirements in separate threat models, and extends the schemes to support additional search semantics to improve search experiences. Analysis demonstrates the schemes provide privacy and efficiency guarantees with low computational and communication overhead.
IRJET- Compound Keyword Search of Encrypted Cloud Data by using Semantic SchemeIRJET Journal
This document proposes a semantic-based compound keyword search (SCKS) scheme for encrypted cloud data. SCKS aims to address limitations of existing approaches by enabling semantic-based, multi-keyword and ranked searches without relying on a predefined global dictionary. It introduces a compound concept semantic similarity calculation method and combines it with Locality-Sensitive Hashing and a secure k-Nearest Neighbor scheme to map keyword vectors to indexes while considering frequency. Experimental results on real-world datasets show SCKS has low computation overhead and higher search accuracy than existing schemes.
We are providing training on IEEE 2016-17 projects for Ph.D Scalars, M.Tech, B.E, MCA, BCA and Diploma students for
all branches for their academic projects.
For more details call us or watsapp us @ 7676768124 0r 9545252155
Email your base papers to "adritsolutions@gmail.co.in"
We are providing IEEE projects on
1) Cloud Computing, Data Mining, BigData Projects Using JAva
2) Image Processing and Video Procesing (MATLAB) , Signal Processing
3) NS2 (Wireless Sensor, MANET, VANET)
4) ANDRIOD APPS
5) JAVA, JEE, J2EE, J2ME
6) Mechanical Design projects
7) Embedded Systems and IoT Projects
8) VLSI- Verilog Projects (ModelSim and Xilinx using FPGA)
For More details Please Visit us at
Adrit Solutions
Near Maruthi Mandir
#42/5, 18th Cross, 21st Main
Vijaynagar
Bangalore.
IRJET - K-Gram based Composite Secret Sign Search Over Encrypted Cloud In...IRJET Journal
This document presents a proposed system called K-Gram based Composite Secret Sign Search over Encrypted Cloud Information. The system aims to provide a qualified search scheme for cloud storage data using multi-keyword search. It uses fuzzy keyword sets to account for spelling errors when searching encrypted file names in the cloud server. If a search keyword matches, related fuzzy keywords are used to search file lists. Experimental results demonstrate that the proposed solution can greatly boost privacy protection, scalability and query processing time efficiency over existing methods. The system considers keyword frequency when ranking search results to better protect user data privacy. It consists of modules for user login, file upload, frequent keyword search, similarity search, linear search, email alerts and file downloading.
Privacy preserving multi-keyword ranked search over encrypted cloud dataIGEEKS TECHNOLOGIES
This document proposes a system called privacy-preserving multi-keyword ranked search over encrypted cloud data (MRSE). Existing searchable encryption systems only support single-keyword or boolean keyword search without result ranking. The proposed MRSE system allows multi-keyword searches over encrypted cloud data and ranks the results based on relevance to the search keywords. It establishes privacy requirements and uses an efficient "coordinate matching" semantic to capture document relevance. The system architecture includes modules for data users, owners, file upload/download and rank search over encrypted data.
International Journal of Engineering Research and DevelopmentIJERD Editor
Electrical, Electronics and Computer Engineering,
Information Engineering and Technology,
Mechanical, Industrial and Manufacturing Engineering,
Automation and Mechatronics Engineering,
Material and Chemical Engineering,
Civil and Architecture Engineering,
Biotechnology and Bio Engineering,
Environmental Engineering,
Petroleum and Mining Engineering,
Marine and Agriculture engineering,
Aerospace Engineering.
Privacy Preserving Ranked Multi-Keyword Search for Multiple Data Owners in Cl...1crore projects
IEEE PROJECTS 2015
1 crore projects is a leading Guide for ieee Projects and real time projects Works Provider.
It has been provided Lot of Guidance for Thousands of Students & made them more beneficial in all Technology Training.
Dot Net
DOTNET Project Domain list 2015
1. IEEE based on datamining and knowledge engineering
2. IEEE based on mobile computing
3. IEEE based on networking
4. IEEE based on Image processing
5. IEEE based on Multimedia
6. IEEE based on Network security
7. IEEE based on parallel and distributed systems
Java Project Domain list 2015
1. IEEE based on datamining and knowledge engineering
2. IEEE based on mobile computing
3. IEEE based on networking
4. IEEE based on Image processing
5. IEEE based on Multimedia
6. IEEE based on Network security
7. IEEE based on parallel and distributed systems
ECE IEEE Projects 2015
1. Matlab project
2. Ns2 project
3. Embedded project
4. Robotics project
Eligibility
Final Year students of
1. BSc (C.S)
2. BCA/B.E(C.S)
3. B.Tech IT
4. BE (C.S)
5. MSc (C.S)
6. MSc (IT)
7. MCA
8. MS (IT)
9. ME(ALL)
10. BE(ECE)(EEE)(E&I)
TECHNOLOGY USED AND FOR TRAINING IN
1. DOT NET
2. C sharp
3. ASP
4. VB
5. SQL SERVER
6. JAVA
7. J2EE
8. STRINGS
9. ORACLE
10. VB dotNET
11. EMBEDDED
12. MAT LAB
13. LAB VIEW
14. Multi Sim
CONTACT US
1 CRORE PROJECTS
Door No: 214/215,2nd Floor,
No. 172, Raahat Plaza, (Shopping Mall) ,Arcot Road, Vadapalani, Chennai,
Tamin Nadu, INDIA - 600 026
Email id: 1croreprojects@gmail.com
website:1croreprojects.com
Phone : +91 97518 00789 / +91 72999 51536
Privacy-Preserving Multi-keyword Top-k Similarity Search Over Encrypted DataCloudTechnologies
Privacy-Preserving Multi-keyword Top-k Similarity Search Over Encrypted Data M-Tech IEEE 2017 Projects B-Tech IEEE 2017 Projects Paper Publishing UGC Journals
IRJET- Multi-Owner Keyword Search over Cloud with CryptographyIRJET Journal
This document discusses multi-owner keyword search over encrypted data stored in the cloud. It notes that most existing schemes only consider a single data owner, and require secure channels to transmit secret keys. It proposes that allowing multiple data owners to share encrypted data and enable keyword searches across this shared data addresses these limitations. The document outlines challenges like ensuring data privacy and confidentiality when outsourcing encrypted data to remote cloud servers. It reviews related work on searchable encryption schemes and their limitations in multi-owner settings. The goal is to enable efficient keyword searches across large encrypted datasets from multiple data owners stored in the cloud.
A Secure and Dynamic Multi-keyword Ranked Search Scheme over Encrypted Cloud ...1crore projects
IEEE PROJECTS 2015
1 crore projects is a leading Guide for ieee Projects and real time projects Works Provider.
It has been provided Lot of Guidance for Thousands of Students & made them more beneficial in all Technology Training.
Dot Net
DOTNET Project Domain list 2015
1. IEEE based on datamining and knowledge engineering
2. IEEE based on mobile computing
3. IEEE based on networking
4. IEEE based on Image processing
5. IEEE based on Multimedia
6. IEEE based on Network security
7. IEEE based on parallel and distributed systems
Java Project Domain list 2015
1. IEEE based on datamining and knowledge engineering
2. IEEE based on mobile computing
3. IEEE based on networking
4. IEEE based on Image processing
5. IEEE based on Multimedia
6. IEEE based on Network security
7. IEEE based on parallel and distributed systems
ECE IEEE Projects 2015
1. Matlab project
2. Ns2 project
3. Embedded project
4. Robotics project
Eligibility
Final Year students of
1. BSc (C.S)
2. BCA/B.E(C.S)
3. B.Tech IT
4. BE (C.S)
5. MSc (C.S)
6. MSc (IT)
7. MCA
8. MS (IT)
9. ME(ALL)
10. BE(ECE)(EEE)(E&I)
TECHNOLOGY USED AND FOR TRAINING IN
1. DOT NET
2. C sharp
3. ASP
4. VB
5. SQL SERVER
6. JAVA
7. J2EE
8. STRINGS
9. ORACLE
10. VB dotNET
11. EMBEDDED
12. MAT LAB
13. LAB VIEW
14. Multi Sim
CONTACT US
1 CRORE PROJECTS
Door No: 214/215,2nd Floor,
No. 172, Raahat Plaza, (Shopping Mall) ,Arcot Road, Vadapalani, Chennai,
Tamin Nadu, INDIA - 600 026
Email id: 1croreprojects@gmail.com
website:1croreprojects.com
Phone : +91 97518 00789 / +91 72999 51536
A Secure and Dynamic Multi-keyword Ranked Search Scheme over Encrypted Cloud ...1crore projects
IEEE PROJECTS 2015
1 crore projects is a leading Guide for ieee Projects and real time projects Works Provider.
It has been provided Lot of Guidance for Thousands of Students & made them more beneficial in all Technology Training.
Dot Net
DOTNET Project Domain list 2015
1. IEEE based on datamining and knowledge engineering
2. IEEE based on mobile computing
3. IEEE based on networking
4. IEEE based on Image processing
5. IEEE based on Multimedia
6. IEEE based on Network security
7. IEEE based on parallel and distributed systems
Java Project Domain list 2015
1. IEEE based on datamining and knowledge engineering
2. IEEE based on mobile computing
3. IEEE based on networking
4. IEEE based on Image processing
5. IEEE based on Multimedia
6. IEEE based on Network security
7. IEEE based on parallel and distributed systems
ECE IEEE Projects 2015
1. Matlab project
2. Ns2 project
3. Embedded project
4. Robotics project
Eligibility
Final Year students of
1. BSc (C.S)
2. BCA/B.E(C.S)
3. B.Tech IT
4. BE (C.S)
5. MSc (C.S)
6. MSc (IT)
7. MCA
8. MS (IT)
9. ME(ALL)
10. BE(ECE)(EEE)(E&I)
TECHNOLOGY USED AND FOR TRAINING IN
1. DOT NET
2. C sharp
3. ASP
4. VB
5. SQL SERVER
6. JAVA
7. J2EE
8. STRINGS
9. ORACLE
10. VB dotNET
11. EMBEDDED
12. MAT LAB
13. LAB VIEW
14. Multi Sim
CONTACT US
1 CRORE PROJECTS
Door No: 214/215,2nd Floor,
No. 172, Raahat Plaza, (Shopping Mall) ,Arcot Road, Vadapalani, Chennai,
Tamin Nadu, INDIA - 600 026
Email id: 1croreprojects@gmail.com
website:1croreprojects.com
Phone : +91 97518 00789 / +91 72999 51536
IRJET- A Survey on Searching of Keyword on Encrypted Data in Cloud using ...IRJET Journal
The document describes a survey on searching for keywords on encrypted data stored in the cloud using an access structure. It discusses how cloud computing allows for large-scale data outsourcing but encrypted data is difficult to search. The paper proposes a method to search encrypted data using access structures expressed as Boolean predicates. Key algorithms discussed are ranked serial binary search to reduce search time and AES encryption to encrypt data and avoid duplication. The goal is to enable efficient keyword searches on encrypted cloud data.
IRJET - Efficient and Verifiable Queries over Encrypted Data in CloudIRJET Journal
This document proposes a scheme for efficient and verifiable queries over encrypted data stored in the cloud. It aims to allow an authorized user to query encrypted documents of interest while maintaining privacy. The scheme provides a verification mechanism to allow users to check the correctness of query results and identify any valid results omitted by a potentially untrustworthy cloud server. The document reviews related work on searchable encryption and verifiable queries. It then outlines the proposed approach to build secure verifiable queries for encrypted cloud data.
This document describes a proposed system for enabling effective yet privacy-preserving fuzzy keyword search in cloud computing. It formalizes the problem of fuzzy keyword search over encrypted cloud data for the first time. The system uses edit distance to quantify keyword similarity and develops two techniques - wildcard-based and gram-based - to construct efficient fuzzy keyword sets. It then proposes a symbol-based trie-traverse searching scheme to match keywords and retrieve files. Security analysis shows the solution preserves privacy while allowing fuzzy searches.
ACCESSING SECURED DATA IN CLOUD COMPUTING ENVIRONMENTIJNSA Journal
Number of businesses using cloud computing has increased dramatically over the last few years due to the attractive features such as scalability, flexibility, fast start-up and low costs. Services provided over the web are ranging from using provider’s software and hardware to managing security and other issues. Some of the biggest challenges at this point are providing privacy and data security to subscribers of public cloud servers. An efficient encryption technique presented in this paper can be used for secure access to and storage of data on public cloud server, moving and searching encrypted data through communication channels while protecting data confidentiality. This method ensures data protection against both external and internal intruders. Data can be decrypted only with the provided by the data owner key, while public cloud server is unable to read encrypted data or queries. Answering a query does not depend on it size and done in a constant time. Data access is managed by the data owner. The proposed schema allows unauthorized modifications detection.
Similar to Ieeepro techno solutions ieee dotnet project - privacy-preserving multi-keyword ranked search (20)
Advanced control scheme of doubly fed induction generator for wind turbine us...IJECEIAES
This paper describes a speed control device for generating electrical energy on an electricity network based on the doubly fed induction generator (DFIG) used for wind power conversion systems. At first, a double-fed induction generator model was constructed. A control law is formulated to govern the flow of energy between the stator of a DFIG and the energy network using three types of controllers: proportional integral (PI), sliding mode controller (SMC) and second order sliding mode controller (SOSMC). Their different results in terms of power reference tracking, reaction to unexpected speed fluctuations, sensitivity to perturbations, and resilience against machine parameter alterations are compared. MATLAB/Simulink was used to conduct the simulations for the preceding study. Multiple simulations have shown very satisfying results, and the investigations demonstrate the efficacy and power-enhancing capabilities of the suggested control system.
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEMHODECEDSIET
Time Division Multiplexing (TDM) is a method of transmitting multiple signals over a single communication channel by dividing the signal into many segments, each having a very short duration of time. These time slots are then allocated to different data streams, allowing multiple signals to share the same transmission medium efficiently. TDM is widely used in telecommunications and data communication systems.
### How TDM Works
1. **Time Slots Allocation**: The core principle of TDM is to assign distinct time slots to each signal. During each time slot, the respective signal is transmitted, and then the process repeats cyclically. For example, if there are four signals to be transmitted, the TDM cycle will divide time into four slots, each assigned to one signal.
2. **Synchronization**: Synchronization is crucial in TDM systems to ensure that the signals are correctly aligned with their respective time slots. Both the transmitter and receiver must be synchronized to avoid any overlap or loss of data. This synchronization is typically maintained by a clock signal that ensures time slots are accurately aligned.
3. **Frame Structure**: TDM data is organized into frames, where each frame consists of a set of time slots. Each frame is repeated at regular intervals, ensuring continuous transmission of data streams. The frame structure helps in managing the data streams and maintaining the synchronization between the transmitter and receiver.
4. **Multiplexer and Demultiplexer**: At the transmitting end, a multiplexer combines multiple input signals into a single composite signal by assigning each signal to a specific time slot. At the receiving end, a demultiplexer separates the composite signal back into individual signals based on their respective time slots.
### Types of TDM
1. **Synchronous TDM**: In synchronous TDM, time slots are pre-assigned to each signal, regardless of whether the signal has data to transmit or not. This can lead to inefficiencies if some time slots remain empty due to the absence of data.
2. **Asynchronous TDM (or Statistical TDM)**: Asynchronous TDM addresses the inefficiencies of synchronous TDM by allocating time slots dynamically based on the presence of data. Time slots are assigned only when there is data to transmit, which optimizes the use of the communication channel.
### Applications of TDM
- **Telecommunications**: TDM is extensively used in telecommunication systems, such as in T1 and E1 lines, where multiple telephone calls are transmitted over a single line by assigning each call to a specific time slot.
- **Digital Audio and Video Broadcasting**: TDM is used in broadcasting systems to transmit multiple audio or video streams over a single channel, ensuring efficient use of bandwidth.
- **Computer Networks**: TDM is used in network protocols and systems to manage the transmission of data from multiple sources over a single network medium.
### Advantages of TDM
- **Efficient Use of Bandwidth**: TDM all
International Conference on NLP, Artificial Intelligence, Machine Learning an...gerogepatton
International Conference on NLP, Artificial Intelligence, Machine Learning and Applications (NLAIM 2024) offers a premier global platform for exchanging insights and findings in the theory, methodology, and applications of NLP, Artificial Intelligence, Machine Learning, and their applications. The conference seeks substantial contributions across all key domains of NLP, Artificial Intelligence, Machine Learning, and their practical applications, aiming to foster both theoretical advancements and real-world implementations. With a focus on facilitating collaboration between researchers and practitioners from academia and industry, the conference serves as a nexus for sharing the latest developments in the field.
Embedded machine learning-based road conditions and driving behavior monitoringIJECEIAES
Car accident rates have increased in recent years, resulting in losses in human lives, properties, and other financial costs. An embedded machine learning-based system is developed to address this critical issue. The system can monitor road conditions, detect driving patterns, and identify aggressive driving behaviors. The system is based on neural networks trained on a comprehensive dataset of driving events, driving styles, and road conditions. The system effectively detects potential risks and helps mitigate the frequency and impact of accidents. The primary goal is to ensure the safety of drivers and vehicles. Collecting data involved gathering information on three key road events: normal street and normal drive, speed bumps, circular yellow speed bumps, and three aggressive driving actions: sudden start, sudden stop, and sudden entry. The gathered data is processed and analyzed using a machine learning system designed for limited power and memory devices. The developed system resulted in 91.9% accuracy, 93.6% precision, and 92% recall. The achieved inference time on an Arduino Nano 33 BLE Sense with a 32-bit CPU running at 64 MHz is 34 ms and requires 2.6 kB peak RAM and 139.9 kB program flash memory, making it suitable for resource-constrained embedded systems.
Using recycled concrete aggregates (RCA) for pavements is crucial to achieving sustainability. Implementing RCA for new pavement can minimize carbon footprint, conserve natural resources, reduce harmful emissions, and lower life cycle costs. Compared to natural aggregate (NA), RCA pavement has fewer comprehensive studies and sustainability assessments.
Redefining brain tumor segmentation: a cutting-edge convolutional neural netw...IJECEIAES
Medical image analysis has witnessed significant advancements with deep learning techniques. In the domain of brain tumor segmentation, the ability to
precisely delineate tumor boundaries from magnetic resonance imaging (MRI)
scans holds profound implications for diagnosis. This study presents an ensemble convolutional neural network (CNN) with transfer learning, integrating
the state-of-the-art Deeplabv3+ architecture with the ResNet18 backbone. The
model is rigorously trained and evaluated, exhibiting remarkable performance
metrics, including an impressive global accuracy of 99.286%, a high-class accuracy of 82.191%, a mean intersection over union (IoU) of 79.900%, a weighted
IoU of 98.620%, and a Boundary F1 (BF) score of 83.303%. Notably, a detailed comparative analysis with existing methods showcases the superiority of
our proposed model. These findings underscore the model’s competence in precise brain tumor localization, underscoring its potential to revolutionize medical
image analysis and enhance healthcare outcomes. This research paves the way
for future exploration and optimization of advanced CNN models in medical
imaging, emphasizing addressing false positives and resource efficiency.
Comparative analysis between traditional aquaponics and reconstructed aquapon...bijceesjournal
The aquaponic system of planting is a method that does not require soil usage. It is a method that only needs water, fish, lava rocks (a substitute for soil), and plants. Aquaponic systems are sustainable and environmentally friendly. Its use not only helps to plant in small spaces but also helps reduce artificial chemical use and minimizes excess water use, as aquaponics consumes 90% less water than soil-based gardening. The study applied a descriptive and experimental design to assess and compare conventional and reconstructed aquaponic methods for reproducing tomatoes. The researchers created an observation checklist to determine the significant factors of the study. The study aims to determine the significant difference between traditional aquaponics and reconstructed aquaponics systems propagating tomatoes in terms of height, weight, girth, and number of fruits. The reconstructed aquaponics system’s higher growth yield results in a much more nourished crop than the traditional aquaponics system. It is superior in its number of fruits, height, weight, and girth measurement. Moreover, the reconstructed aquaponics system is proven to eliminate all the hindrances present in the traditional aquaponics system, which are overcrowding of fish, algae growth, pest problems, contaminated water, and dead fish.
A SYSTEMATIC RISK ASSESSMENT APPROACH FOR SECURING THE SMART IRRIGATION SYSTEMSIJNSA Journal
The smart irrigation system represents an innovative approach to optimize water usage in agricultural and landscaping practices. The integration of cutting-edge technologies, including sensors, actuators, and data analysis, empowers this system to provide accurate monitoring and control of irrigation processes by leveraging real-time environmental conditions. The main objective of a smart irrigation system is to optimize water efficiency, minimize expenses, and foster the adoption of sustainable water management methods. This paper conducts a systematic risk assessment by exploring the key components/assets and their functionalities in the smart irrigation system. The crucial role of sensors in gathering data on soil moisture, weather patterns, and plant well-being is emphasized in this system. These sensors enable intelligent decision-making in irrigation scheduling and water distribution, leading to enhanced water efficiency and sustainable water management practices. Actuators enable automated control of irrigation devices, ensuring precise and targeted water delivery to plants. Additionally, the paper addresses the potential threat and vulnerabilities associated with smart irrigation systems. It discusses limitations of the system, such as power constraints and computational capabilities, and calculates the potential security risks. The paper suggests possible risk treatment methods for effective secure system operation. In conclusion, the paper emphasizes the significant benefits of implementing smart irrigation systems, including improved water conservation, increased crop yield, and reduced environmental impact. Additionally, based on the security analysis conducted, the paper recommends the implementation of countermeasures and security approaches to address vulnerabilities and ensure the integrity and reliability of the system. By incorporating these measures, smart irrigation technology can revolutionize water management practices in agriculture, promoting sustainability, resource efficiency, and safeguarding against potential security threats.
1. Privacy-Preserving Multi-keyword Ranked Search
over Encrypted Cloud Data
Ning Cao†, Cong Wang‡, Ming Li†, Kui Ren‡, and Wenjing Lou†
†Department of ECE, Worcester Polytechnic Institute, Email: {ncao, mingli, wjlou}@ece.wpi.edu
‡Department of ECE, Illinois Institute of Technology, Email: {cong, kren}@ece.iit.edu
Abstract—With the advent of cloud computing, data owners
are motivated to outsource their complex data management
systems from local sites to commercial public cloud for great
flexibility and economic savings. But for protecting data privacy,
sensitive data has to be encrypted before outsourcing, which
obsoletes traditional data utilization based on plaintext keyword
search. Thus, enabling an encrypted cloud data search service is
of paramount importance. Considering the large number of data
users and documents in cloud, it is crucial for the search service to
allow multi-keyword query and provide result similarity ranking
to meet the effective data retrieval need. Related works on
searchable encryption focus on single keyword search or Boolean
keyword search, and rarely differentiate the search results. In
this paper, for the first time, we define and solve the challenging
problem of privacy-preserving multi-keyword ranked search over
encrypted cloud data (MRSE), and establish a set of strict privacy
requirements for such a secure cloud data utilization system
to become a reality. Among various multi-keyword semantics,
we choose the efficient principle of “coordinate matching”, i.e.,
as many matches as possible, to capture the similarity between
search query and data documents, and further use “inner product
similarity” to quantitatively formalize such principle for similar-
ity measurement. We first propose a basic MRSE scheme using
secure inner product computation, and then significantly improve
it to meet different privacy requirements in two levels of threat
models. Thorough analysis investigating privacy and efficiency
guarantees of proposed schemes is given, and experiments on
the real-world dataset further show proposed schemes indeed
introduce low overhead on computation and communication.
I. INTRODUCTION
Cloud computing is the long dreamed vision of computing
as a utility, where cloud customers can remotely store their
data into the cloud so as to enjoy the on-demand high quality
applications and services from a shared pool of configurable
computing resources [1]. Its great flexibility and economic
savings are motivating both individuals and enterprises to
outsource their local complex data management system into
the cloud, especially when the data produced by them that need
to be stored and utilized is rapidly increasing. To protect data
privacy and combat unsolicited accesses in cloud and beyond,
sensitive data, e.g., emails, personal health records, photo al-
bums, tax documents, financial transactions, etc., may have to
be encrypted by data owners before outsourcing to commercial
public cloud [2]; this, however, obsoletes the traditional data
utilization service based on plaintext keyword search. The
trivial solution of downloading all the data and decrypting
locally is clearly impractical, due to the huge amount of
bandwidth cost in cloud scale systems. Moreover, aside from
eliminating the local storage management, storing data into
the cloud serves no purpose unless they can be easily searched
and utilized. Thus, exploring privacy-preserving and effective
search service over encrypted cloud data is of paramount
importance. Considering the potentially large number of on-
demand data users and huge amount of outsourced data
documents in cloud, this problem is particularly challenging
as it is extremely difficult to meet also the requirements of
performance, system usability and scalability.
On the one hand, to meet the effective data retrieval need,
large amount of documents demand cloud server to perform
result relevance ranking, instead of returning undifferenti-
ated result. Such ranked search system enables data users
to find the most relevant information quickly, rather than
burdensomely sorting through every match in the content
collection [3]. Ranked search can also elegantly eliminate
unnecessary network traffic by sending back only the most
relevant data, which is highly desirable in the “pay-as-you-
use” cloud paradigm. For privacy protection, such ranking
operation, however, should not leak any keyword related
information. On the other hand, to improve search result
accuracy as well as enhance user searching experience, it
is also crucial for such ranking system to support multiple
keywords search, as single keyword search often yields far too
coarse result. As a common practice indicated by today’s web
search engines (e.g., Google search), data users may tend to
provide a set of keywords instead of only one as the indicator
of their search interest to retrieve the most relevant data. And
each keyword in the search request is able to help narrow down
the search result further. “Coordinate matching” [4], i.e., as
many matches as possible, is an efficient principle among such
multi-keyword semantics to refine the result relevance, and has
been widely used in the plaintext information retrieval (IR)
community. However, how to apply it in the encrypted cloud
data search system remains a very challenging task because
of inherent security and privacy obstacles, including various
strict requirements like data privacy, index privacy, keyword
privacy, and many others (see section III-B).
In the literature, searchable encryption [5]–[13] is a helpful
technique that treats encrypted data as documents and allows
a user to securely search over it through single keyword and
retrieve documents of interest. However, direct application
of these approaches to deploy secure large scale cloud data
utilization system would not be necessarily suitable, as they
are developed as crypto primitives and cannot accommodate
such high service-level requirements like system usability, user
searching experience, and easy information discovery in mind.
2. 2
Although some recent designs have been proposed to support
Boolean keyword search [14]–[21] as an attempt to enrich
the search flexibility, they are still not adequate to provide
users with acceptable result ranking functionality (see section
VI). Our early work [22] has been aware of this problem,
and solves the secure ranked search over encrypted data with
support of only single keyword query. But how to design an
efficient encrypted data search mechanism that supports multi-
keyword semantics without privacy breaches still remains an
challenging open problem.
In this paper, for the first time, we define and solve the
problem of multi-keyword ranked search over encrypted cloud
data (MRSE) while preserving strict system-wise privacy in
cloud computing paradigm. Among various multi-keyword
semantics, we choose the efficient principle of “coordinate
matching”, i.e., as many matches as possible, to capture
the similarity between search query and data documents.
Specifically, we use “inner product similarity” [4], i.e., the
number of query keywords appearing in a document, to
quantitatively evaluate the similarity of that document to the
search query in “coordinate matching” principle. During index
construction, each document is associated with a binary vector
as a subindex where each bit represents whether corresponding
keyword is contained in the document. The search query is also
described as a binary vector where each bit means whether
corresponding keyword appears in this search request, so the
similarity could be exactly measured by inner product of query
vector with data vector. However, directly outsourcing data
vector or query vector will violate index privacy or search
privacy. To meet the challenge of supporting such multi-
keyword semantic without privacy breaches, we propose a
basic MRSE scheme using secure inner product computation,
which is adapted from a secure k-nearest neighbor (kNN)
technique [4], and then improve it step by step to achieve
various privacy requirements in two levels of threat models.
Our contributions are summarized as follows,
1) For the first time, we explore the problem of multi-
keyword ranked search over encrypted cloud data, and
establish a set of strict privacy requirements for such a
secure cloud data utilization system to become a reality.
2) We propose two MRSE schemes following the principle
of “coordinate matching” while meeting different pri-
vacy requirements in two levels of threat models.
3) Thorough analysis investigating privacy and efficiency
guarantees of proposed schemes is given, and experi-
ments on the real-world dataset further show proposed
schemes indeed introduce low overhead on computation
and communication.
The remainder of this paper is organized as follows. In Sec-
tion II, we introduce the system and threat model, our design
goals, and preliminary. Section III describes MRSE framework
and privacy requirements, followed by section IV, which gives
our schemes achieving efficiency and privacy requirements.
Section V presents simulation results. We discuss related work
on both single and Boolean keyword searchable encryption in
Section VI, and conclude the paper in Section VII.
data & indexencrypted
search request
Data owner
Semi-trusted
cloud server
ranked result
Data users
search control(trapdoors)
access control(data decryption keys)
Fig. 1: Architecture of the search over encrypted cloud data
II. PROBLEM FORMULATION
A. System Model
Considering a cloud data hosting service involving three
different entities, as illustrated in Fig. 1: data owner, data
user, and cloud server. Data owner has a collection of data
documents F to be outsourced to cloud server in the encrypted
form C. To enable the searching capability over C for effective
data utilization, data owner, before outsourcing, will first build
an encrypted searchable index I from F, and then outsource
both the index I and the encrypted document collection C
to cloud server. To search the document collection for t
given keywords, an authorized user acquires a corresponding
trapdoor T through search control mechanisms, e.g., broadcast
encryption [8]. Upon receiving T from data users, cloud server
is responsible to search the index I and return the corre-
sponding set of encrypted documents. To improve document
retrieval accuracy, search result should be ranked by cloud
server according to some ranking criteria (e.g., coordinate
matching, as will be introduced shortly). Moreover, to reduce
communication cost, data user may send an optional number
k along with the trapdoor T so that cloud server only sends
back top-k documents that are most relevant to the search
query. Finally, the access control mechanism is employed to
manage decryption capabilities given to users.
B. Threat Model
Cloud server is considered as “honest-but-curious” in our
model, which is consistent with the most related works on
searchable encryption. Specifically, cloud server acts in an
“honest” fashion and correctly follows the designated protocol
specification. However, it is “curious” to infer and analyze data
(including index) in its storage and message flows received
during the protocol so as to learn additional information. Based
on what information cloud server knows, we consider two
levels of threat models as follows.
Known Ciphertext Model In this model, cloud server is
supposed to only know encrypted dataset C and searchable
index I, both of which are outsourced from data owner.
Known Background Model In this stronger model, cloud
server is supposed to possess some backgrounds on the dataset,
such as the subject and its related statistical information, in
addition to what can be accessed in known ciphertext model.
As an instance of possible attacks in this case, cloud server
could utilize document frequency or keyword frequency [23]
to identify keywords in the query.
3. 3
C. Design Goals
To enable ranked search for effective utilization of out-
sourced cloud data under the aforementioned model, our
system design should simultaneously achieve security and
performance guarantees as follows.
• Multi-keyword Ranked Search: To design search
schemes which allow multi-keyword query and provide
result similarity ranking for effective data retrieval, in-
stead of returning undifferentiated results.
• Privacy-Preserving: To prevent cloud server from learn-
ing additional information from dataset and index, and to
meet privacy requirements specified in section III-B.
• Efficiency: Above goals on functionality and privacy
should be achieved with low communication and com-
putation overhead.
D. Notations
• F – the plaintext document collection, denoted as a set
of m data documents F = (F1, F2, . . . , Fm).
• C – the encrypted document collection stored in cloud
server, denoted as C = (C1, C2, . . . , Cm).
• W – the distinct keywords extracted from document
collection F, denoted as W = (W1, W2, . . . , Wn).
• I – the searchable index associated with C, denoted as
(I1, I2, . . . , Im) where each subindex Ii is built for Fi.
• W – the subset of W, representing the keywords in a
search request, denoted as W = (Wj1
, Wj2
, . . . , Wjt
).
• TW
– the trapdoor for the search request W.
• FW
– the ranked id list of all documents according to
their similarity with W.
E. Preliminary on Coordinate Matching
As a hybrid of conjunctive search and disjunctive search,
“coordinate matching” [4] is an intermediate approach which
uses the number of query keywords appearing in the document
to quantify the similarity of that document to the query. When
users know the exact subset of the dataset to be retrieved,
Boolean queries perform well with the precise search require-
ment specified by the user. In cloud computing, however, this
is not the practical case, given the huge amount of outsourced
data. Therefore, it is more flexible for users to specify a list
of keywords indicating their interest and retrieve the most
relevant documents with rank order.
III. FRAMEWORK AND PRIVACY REQUIREMENTS FOR
MRSE
In this section, we define the framework of multi-keyword
ranked search over encrypted cloud data (MRSE) and establish
various strict system-wise privacy requirements for such a
secure cloud data utilization system.
A. MRSE Framework
For easy presentation, operations on the data documents are
not shown in the framework since data owner could easily
employ traditional symmetric key cryptography to encrypt and
then outsource data. With focus on index and query, a MRSE
consists of four algorithms as follows.
• Setup(1 ) Taking a security parameter as input, data
owner outputs a symmetric key as SK.
• BuildIndex(F, SK) Based on the dataset F, data owner
builds a searchable index I which is encrypted by the
symmetric key SK and then outsourced to cloud server.
After the index construction, the document collection can
be independently encrypted and outsourced.
• Trapdoor(W) With t keywords of interest in W as input,
this algorithm generates a corresponding trapdoor TW
.
• Query(TW
, k, I) When cloud server receives a query
request as (TW
, k), it performs the ranked search on
the index I with the help of trapdoor TW
, and finally
returns FW
, the ranked id list of top-k documents sorted
by their similarity with W.
Both search control and access control are not within the
scope of this paper. While the former is to regulate how
authorized users acquire trapdoors, the later is to manage
users’ access to outsourced documents.
B. Privacy Requirements for MRSE
The representative privacy guarantee in the related litera-
ture, such as searchable encryption, is that the server should
learn nothing but search results. With this general privacy
description, we explore and establish a set of strict privacy
requirements specifically for the MRSE framework.
As for the data privacy, data owner can resort to traditional
symmetric key cryptography to encrypt the data before out-
sourcing, and successfully prevent cloud server from prying
into outsourced data. With respect to the index privacy, if
server deduces any association between keywords and en-
crypted documents from index, it may learn the major subject
of a document, even the content of a short document [23].
Therefore, searchable index should be constructed to prevent
server from performing such kind of association attack. While
data and index privacy guarantees are demanded by default
in the related literature, various search privacy requirements
involved in the query procedure are more complex and difficult
to tackle as follows.
Keyword Privacy As users usually prefer to keep their search
from being exposed to others like cloud server, the most
important concern is to hide what they are searching, i.e., the
keywords indicated by the corresponding trapdoor. Although
the trapdoor can be generated in a cryptographic way to protect
the query keywords, cloud server could do some statistical
analysis over the search result to make an estimate. As a
kind of statistical information, document frequency (i.e., the
number of documents containing the keyword) is sufficient to
identify the keyword with high probability [24]. When cloud
server knows some background information of the dataset,
this keyword specific information may be utilized to reverse-
engineer the keyword.
Trapdoor Privacy Since only authorized users are allowed
to acquire trapdoors for their search query, the server is not
4. 4
expected to have the ability to generate valid trapdoors from
previous received ones. Specifically, given one trapdoor for a
set of multiple keywords, the server is not allowed to generate
a valid trapdoor for its subset, including single keyword.
For example, it is forbidden to generate or deduce a new
trapdoor as TWi for keyword Wi from the received trapdoor as
T(Wi,Wk) for two keywords (Wi, Wj). Moreover, the server is
not allowed to generate a valid trapdoor, e.g., T(Wi,Wj ), from
two or more trapdoors, like T(Wi,Wk) and T(Wj ,Wk).
Search Pattern In accordance with the definition in related
work on single keyword searchable encryption [8], search
pattern of data user in MRSE means any information that can
be derived by server if it acquires the knowledge that two
arbitrary searches are performed for the same keywords or
not. If the trapdoor is generated in a deterministic manner,
server could easily know the search pattern of any data
user by comparing trapdoors received from that user. So
the fundamental protection for search pattern is to introduce
nondeterminacy into trapdoor generation procedure.
Access Pattern Within the ranked search, access pattern is
the sequence of search results where every search result is
a set of documents with rank order. Specifically, the search
result for W is denoted as FW
, consisting of the id list of all
documents ranked by their similarity to W. Then the access
pattern is denoted as (FW1
, FW2
, . . .) which are the results
of sequential searches. Although a few searchable encryption
works, e.g., [17] has been proposed to utilize private informa-
tion retrieval(PIR) technique [25], to hide access pattern, our
proposed schemes are not designed to protect access pattern
for the efficiency concerns. This is because any PIR based
technique must “touch” the whole dataset outsourced on the
server which is inefficient in the large scale cloud system.
IV. PRIVACY-PRESERVING AND EFFICIENT MRSE
To efficiently achieve multi-keyword ranked search, we pro-
pose to employ “inner product similarity” [4] to quantitatively
formalize the efficient ranking principle “coordinate match-
ing”. Specifically, Di is a binary data vector for document
Fi where each bit Di[j] ∈ {0, 1} represents the existence
of the corresponding keyword Wj in that document, and Q
is a binary query vector indicating the keywords of interest
where each bit Q[j] ∈ {0, 1} represents the existence of the
corresponding keyword Wj in the query W. The similarity
score of document Fi to query W is therefore expressed as
the inner product of their binary column vectors, i.e., Di · Q.
For the purpose of ranking, cloud server must be given the
capability to compare the similarity of different documents
to the query. But, to preserve strict system-wise privacy, data
vector Di, query vector Q and their inner product Di·Q should
not be exposed to cloud server. In this section, we first propose
a basic MRSE scheme using secure inner product computation,
which is adapted from a secure k-nearest neighbor (kNN)
technique, and then show how to significantly improve it to
be privacy-preserving against different levels of threat models
in the MRSE framework in a step-by-step manner.
A. MRSE I: Basic Scheme
1) Secure kNN Computation: In the secure k-nearest neigh-
bor (kNN) scheme [26], Euclidean distance between a database
record pi and a query vector q is used to select k nearest
database records. The secret key is composed of one (d + 1)-
bit vector as S and two (d + 1) × (d + 1) invertible matrices
as {M1, M2}, where d is the number of fields for each record
pi. First, every data vector pi and query vector q are extended
to (d + 1)-dimension vectors as pi and q, where the (d + 1)-
th dimension is set to −0.5||p2
i || and 1, respectively. Besides,
the query vector q is scaled by a random number r > 0 as
(rq, r). Then, pi is split into two random vectors as {pi , pi },
and q is also split into two random vectors as {q , q }.
Note here that vector S functions as a splitting indicator.
Namely, if the j-th bit of S is 0, pi [j] and pi [j] are set
as the same as pi[j], while q [j] and q [j] are set to two
random numbers so that their sum is equal to q[j]; if the j-
th bit of S is 1, the splitting process is similar except that
pi and q are switched. The split data vector pair {pi , pi } is
encrypted as {MT
1 pi , MT
2 pi }, and the split query vector pair
{q , q } is encrypted as {M−1
1 q , M−1
2 q }. In the query
step, the product of data vector pair and query vector pair, i.e.,
−0.5r(||pi||2
−2pi ·q), is serving as the indicator of Euclidean
distance (||pi||2
−2pi ·q+||q||2
) to select k nearest neighbors.
Without prior knowledge of secret key, neither data vector nor
query vector, after such a series of processes, can be recovered
by analyzing their corresponding ciphertext.
As MRSE is using inner product similarity instead of
Euclidean distance, we need to do some modifications on
the data structure to fit the MRSE framework. By eliminating
dimension extension, the final result changes to be the inner
product as rpi · q, and it seems that an efficient inner product
computation scheme can be directly achieved. However, this
approach hides the product only by a scale factor r which will
leak the relationship among different queries. For example, if
two queries are taking for the same keywords, denoted as rq
and r q, similarity scores in two queries will satisfy the scale
relationship, i.e., (pi·rq)/(pi·r q) = (pj ·rq)/(pj ·r q) = r/r .
As a consequence, the search pattern of data user is leaked
via examining whether similarity scores for all documents in
two queries hold such relationship.
2) MRSE I Scheme: To provide a guarantee against vi-
olation on search pattern clearly presented above, we have
to eliminate the scale relationship among similarity scores
in different queries. To do so, instead of simply removing
the extended dimension as we plan to do at the first glance,
we preserve this dimension extending operation but assign
a random number to the extended dimension in each query
vector. The whole scheme to achieve ranked search with
multiple keywords over encrypted data is as follows.
• Setup After extracting the distinct keywords set W
from the document collection F, data owner randomly
generates a (n+1)-bit vector as S and two (n+1)×(n+1)
invertible matrices {M1, M2}. The secret key SK is in
the form of a 3-tuple as {S, M1, M2}.
• BuildIndex(F, SK) Data owner generates a binary data
vector Di for every document Fi, where each binary bit
5. 5
Di[j] represents the existence of the corresponding key-
word Wj in that document. Subsequently, every subindex
Ii is generated by applying dimension extending, splitting
and encrypting procedures on Di. These procedures are
similar with those above except that the (n + 1)-th entry
in Di is set to 1 during dimension extending. Finally,
subindex Ii = {MT
1 Di , MT
2 Di } is built for every
encrypted document Ci on cloud server.
• Trapdoor(W) With t keywords of interest in W as input,
one binary vector Q is first generated where each bit Q[j]
indicates whether Wj ∈ W is true or false. Q is then
scaled by a random number r = 0 as rQ, and extended
to a (n + 1)-dimension vector as Q = (rQ, t) where t is
another random number. After applying the same splitting
and encrypting processes as above, the trapdoor TW
is
generated as {M−1
1 Q , M−1
2 Q }.
• Query(TW
, k, I) With the trapdoor TW
, cloud server
computes the similarity scores of each document Fi as in
equation 1. WLOG, we assume r > 0. After sorting all
scores, cloud server returns the top-k ranked id list FW
.
With the randomness t brought into the query vector, the
final similarity scores do not keep the proportional relationship
to the original inner product and therefore prevent cloud server
from guessing search pattern through search results.
Ii · TW
= {MT
1 Di , MT
2 Di } · {M−1
1 Q , M−1
2 Q } (1)
= Di · q + Di · q = Di · Q = r(Di · Q) + t.
3) Analysis: We analyze this basic MRSE scheme from
three aspects of design goals described in section II.
Functionality and Efficiency Assume the number of query
keywords appearing in a document Fi is xi = Di · Q. From
equation 1, the final similarity score as yi = Ii ·TW
= rxi +t
is a linear function of xi. Besides, the coefficient r is set as a
positive random number, so the order of similarity is exactly
preserved for all the outsourced documents. When cloud server
creates the FW
by the final similarity score yi, the top-k
most similar documents to the query are included with the
precise rank order. Similar with the secure kNN scheme, our
inner product based MRSE scheme is an outstanding approach
from the performance perspective. In the step like BuildIndex
or Trapdoor, the generation procedure of each subindex or
trapdoor involves two multiplications of a (n + 1) × (n + 1)
matrix and a (n + 1)-dimension vector. In the Query, the
final similarity score is computed through two multiplications
of two (n + 1)-dimension vectors.
Privacy As for the data privacy, traditional symmetric key
encryption techniques could be properly utilized here and
is not within the scope of this paper. The index privacy
is well protected if the secret key SK is kept confidential
since such vector encryption method has been proved to be
secure in known ciphertext model [26]. In addition to the
random number t in the query result, our basic scheme can
generate two totally different trapdoors for the same query W.
Therefore, the search pattern is well protected in our basic
scheme, while it is an unsolved privacy leakage problem in
related symmetric key based searchable encryption schemes
because of the deterministic property of trapdoor generation.
TABLE I: Min/Max Score Analysis I
r > 0 min{yj} = r · min{xi} + t
r > 0 submin{yj} = r · submin{xi} + t
r < 0 max{yj} = r · max{xi} + t
r < 0 submax{yj} = r · submax{xi} + t
B. MRSE II: Privacy-Preserving Scheme in Known Cipher-
text Model
MRSE I scheme performs outstanding from the efficiency
perspective and also provides privacy guarantee on search
pattern, but it will incur trapdoor privacy leakage once cloud
server is requested by users to execute two or more times
of searches. By analyzing similarity scores obtained during
search, cloud server has a chance to deduce a valid trapdoor
which violates trapdoor privacy goal. We first demonstrate how
such analysis attack works, and then show this problem can
be fixed through inserting dummy keyword.
1) Min/Max Score Analysis Attack: With any two valid
trapdoors T1 and T2 submit by users, cloud server may
explore the relationship among similarity scores to deduce
a new trapdoor T3. If T1 and T2 happen to be trapdoors
for two related sets of query keywords, like {K1, K2} and
{K1, K2, K3, K4}, T3 then becomes a valid trapdoor for
keywords {K3, K4}. To illustrate, the relationship between
final similarity score yj with the original one xi is listed
in Tab. I. WLOG, we only discuss the case where r > 0.
In the query Q1 with the trapdoor T1 = {T1[1], T1[2]},
there may exist a document containing neither K1 nor K2
with very high probability. As a result, the minimal original
similarity score as min{xi} is equal to 0 and the minimal
final similarity score as min{yj} becomes t1. Considering
the large number of outsourced documents, it is very likely
that there also exists a document containing only one of the
two keywords, and then the subminimal original similarity
score as submin{xi} is equal to 1 and the subminimal final
similarity score as submin{yj} becomes r1 +t1. With t1 and
r1 + t1, cloud server solves r1 with apparent ease. Similarly,
the other two parameters r2 and t2 can be figured out with
T2 = {T2[1], T2[2]} for query Q2. Assume that the original
query vector for {K3, K4} is denoted as Q3. Note that Q3 is
equal to Q2 − Q1 according to the definition of binary query
vector and the relationship between three keyword sets. Then
the 2-tuple as {T2[1]/r2 − T1[1]/r1, T2[2]/r2 − T1[2]/r1} can
be utilized by cloud server as a valid trapdoor T3 for query
Q3 where r3 is set to 1 and t3 is set to (r2/r2 − t1/r1). The
effectiveness of T3 is validated in equation 2.
Ii · T3
= Ii · {T2[1]/r2 − T1[1]/r1, T2[2]/r2 − T1[2]/r1}
=
Ii[1] · T2[1] + Ii[2] · T2[2]
r2
−
Ii[1] · T1[1] + Ii[2] · T1[2]
r1
= (Di · Q2)/r2 − (Di · Q1)/r1
= (Di · Q2 + t2/r2) − (Di · Q1 + t1/r1)
= r3(Di · Q3) + t3 (2)
Actually, such kind of trapdoor deduction enables cloud server
to perform query that is not requested by any user, and thus
6. 6
TABLE II: Min/Max Score Analysis II
r > 0 min{yj} = r · min{xi + εi} + t
r > 0 submin{yj} = r · submin{xi + εi} + t
r < 0 max{yj} = r · max{xi + εi} + t
r < 0 submax{yj} = r · submax{xi + εi} + t
violates search privacy, specifically the trapdoor privacy.
2) MRSE II Scheme: The previous analysis attack shows
that the trapdoor privacy problem stems from the determin-
istic relationship between the minimal/subminimal (or max-
imal/submaximal) final similarity score and two parameters
as r and t. Therefore, breaking the deterministic relationship
is an alternative to protect trapdoor privacy. In this section,
we propose an improved MRSE scheme preserving trapdoor
privacy in the known ciphertext model. Namely, we will show
how to break the deterministic relationship while maintaining
as much efficiency and accuracy as possible. Introducing some
randomness in the final similarity score is an effective way
towards what we expect here. More specifically, unlike the
randomness involved in the query vector, we insert a dummy
keyword into each data vector and assign a random value to
it. All the vectors are extended to (n + 2)-dimension instead
of (n+1), and each entry in Di is not a binary value anymore
for storing the random variable εi of the dummy keyword in
the (n + 2)-th dimension. Improvement details in MRSE II
scheme is shown as follows.
• Setup Data owner randomly generates a (n+2)-bit vector
as S and two (n + 2) × (n + 2) invertible matrices as
{M1, M2}.
• BuildIndex(F, SK) The (n + 2)-th entry in Di is set to
a random number εi during the dimension extending.
• Trapdoor(W) The (n + 2)-th entry in Q is always set to
1 for any query Q.
• Query(TW
, k, I) The final similarity score computed by
cloud server is equal to r(xi + εi) + ti.
3) Analysis: We follow the same steps as in MRSE I.
Functionality and Efficiency Because the randomness is
introduced as a part of the similarity score, the final search
result on the basis of sorting similarity scores may not be
as accurate as that in MRSE I scheme. For the considera-
tion of search accuracy, let ε follow a normal distribution
N(0, σ2
). To quantitatively evaluate the search accuracy, we
set a measure as precision Pk to capture the fraction of
returned top-k documents that are included in the real top-
k list. Detailed accuracy evaluation on the real-world dataset
will be given in section VI. Besides, MRSE II scheme takes
similar computation and communication costs as in MRSE I
scheme, except that all the vector multiplication operations are
run in the (n + 2)-dimension instead of (n + 1)-dimension.
Privacy We first take a look at trapdoor privacy, especially
the Min/Max score analysis. Similar with the case discussed
above, we assume that cloud server has two trapdoors T1
and T2 for query keywords {K1, K2} and {K1, K2, K3, K4}
respectively. To perform Min/Max score analysis, the server
scans the entire index and computes all the final similarity
scores as listed in Tab. II. With the interference of ε, the
adversary cannot explore useful relationship between the min-
TABLE III: K3 appears in every document
Doc Query for {K1, K2, K3} Query for {K1, K2}
1 x1 = 3, y1 = r(3 + ε1) + t x 1 = 2, y 1 = r (2 + ε1) + t
2 x2 = 2, y2 = r(2 + ε2) + t x 2 = 1, y 2 = r (1 + ε1) + t
3 x3 = 1, y3 = r(1 + ε3) + t x 3 = 0, y 3 = r (0 + ε3) + t
TABLE IV: K3 does not appear in either document
Doc Query for {K1, K2, K3} Query for {K1, K2}
1 x1 = 2, y1 = r(2 + ε1) + t x 1 = 2, y 1 = r (2 + ε1) + t
2 x2 = 1, y2 = r(1 + ε2) + t x 2 = 1, y 2 = r (1 + ε1) + t
3 x3 = 0, y3 = r(0 + ε3) + t x 3 = 0, y 3 = r (0 + ε3) + t
imal/subminimal final similarity score and two parameters as
r and t, and thus the Min/Max score analysis attack does not
work anymore. In addition, any single trapdoor itself does
not reveal valuable information, like the positions of 1 in
its original query vector Q, and therefore keyword privacy is
well protected. To sum up, in the known ciphertext model,
MRSE II scheme meets all expected privacy requirements
mentioned in section III-B.
C. MRSE III: Privacy-Preserving Scheme in Known Back-
ground Model
When cloud server has known some background informa-
tion on the outsourced dataset, keyword privacy cannot be
guaranteed anymore by MRSE II scheme. This is possible in
known background model because cloud server can use scale
analysis as follows to deduce the keyword-specific informa-
tion, e.g., document frequency, which can be further com-
bined with background information to identify the keyword
in a query at high probability. After presenting how cloud
server uses scale analysis attack to break keyword privacy,
we propose a more advanced MRSE scheme to be privacy-
preserving in known background model.
1) Scale Analysis Attack: Given two correlated trapdoors
T1 and T2 for query keywords {K1, K2} and {K1, K2, K3}
respectively, there will be two special cases when searching on
any three documents as listed in Tab. III and Tab. IV. In any
of these two cases, there exists a system of equations among
those similarity scores as follows,
y1 − y2 = r(1 + ε1 − ε2);
y 1 − y 2 = r (1 + ε1 − ε2);
y1 − y3 = r(2 + ε1 − ε2);
y 1 − y 3 = r (2 + ε1 − ε2).
(3)
And cloud server could deduce the following scale relationship
among all the six scores being consistent with equation 3,
(y1 − y2)/(y 1 − y 2) = (y1 − y3)/(y 1 − y 3). (4)
To this end, although the exact value of xi is encrypted as yi,
cloud server, based on the equivalence of (y1 −y2)/(y 1 −y 2)
and (y1 − y3)/(y 1 − y 3), could deduce that whether all the
three documents contain K3 or none of them contain K3. By
extending three documents to the whole dataset, cloud server
could deduce two possible values of document frequency of
the keyword K3. In known background model, the server can
further identify the keyword K3 by referring to the keyword
specific document frequency information about the dataset.
7. 7
50 70 90 110 130 150
50
60
70
80
90
100
# of retrieved documents
Precision(%)
σ = 1
σ = 0.5
(a)
50 70 90 110 130 150
0
10
20
30
40
50
# of retrieved documents
Privacy(%)
σ = 1
σ = 0.5
(b)
Fig. 2: With different choice of standard deviation σ for the
random variable ε, there exists tradeoff between (a) Precision,
and (b) Rank Privacy.
2) MRSE III Scheme: The privacy leakage shown above is
caused by the fixed value of random variable εi in data vector
Di. To eliminate such fixed property in any specific document,
more dummy keywords instead of only one should be inserted
into every data vector Pi. All the vectors are extended to (n+
U + 1)-dimension instead of (n + 2), where U is the number
of dummy keywords inserted. Improved details in MRSE III
scheme is presented as follows.
• Setup(1n
) Data owner randomly generates a (n+U +1)-
bit vector as S and two (n+U +1)×(n+U +1) invertible
matrices {M1, M2}.
• BuildIndex(F, SK) The (n+j+1)-th entry in Di where
j ∈ [1, U] is set to a random number ε(j)
during the
dimension extending.
• Trapdoor(W) By randomly selecting V out of U dummy
keywords, the corresponding entries in Q are set to 1.
• Query(TW
, k, I) The final similarity score computed by
cloud server is equal to r(xi + ε
(v)
i ) + ti where the
v-th dummy keyword is included in the V selected ones.
3) Analysis: To achieve the 80-bit security level, there
should be at least 280
different values of ε
v)
i for each data
vector. The number of different ε
(v)
i is equal to (U
V ), which
is maximized when U
V = 2. Besides, considering (U
V ) ≥ (U
V )V
,
it is greater than 280
when U = 160 and V = 80. So every data
vector should include 160 dummy elements, and every query
vector will randomly select 80 dummy elements. To this end,
MRSE III scheme is secure against scale analysis attack, and
provides various expected privacy guarantees within known
ciphertext model or known background model.
Moreover, to make ε
(v)
i follow the Normal distribution
N(0, σ2
) as above, every ε(j)
is assumed to follow the same
uniform distribution M(−c, c), where the mean as µj is 0
and the variance as σ2
j is c2
/3. According to the central limit
theorem, the sum of 80 independent random variables ε(j)
follows the Normal distribution, where µ = 80µj=0 and σ2
=
80σ2
j = 80c2
/3. Therefore, the value of c is set as 3
80 σ.
With such parameter setting, search accuracy is statistically
the same as that in MRSE II scheme.
1 2 3 4 5 6 7
2
4
6
8
10
12
# of douments in dataset (x 10
2
)
#ofdistinctkeywords(x103
)
(a)
1 2 3 4 5 6 7
0
5
10
15
20
25
# of documents in dataset (x 102
)
Timeofbuildingindex(x102
s)
MRSE_I
MRSE_II
MRSE_III
(b)
Fig. 3: Relationship between number of documents in dataset
and (a) Number of distinct keywords in dataset, and (b) Time
cost for building searchable index.
V. PERFORMANCE ANALYSIS
In this section, we demonstrate a thorough experimental
evaluation of the proposed technique on a real-world dataset:
the Enron Email Dataset [27]. We randomly select different
number of emails to build dataset. The whole experiment
system is implemented by C language on a Linux Server
with Intel Xeon Processor 2.93GHz. The public utility routines
by Numerical Recipes are employed to compute the inverse
of matrix. The performance of our technique is evaluated
regarding the efficiency of three proposed MRSE schemes,
as well as the tradeoff between search precision and privacy.
A. Precision and Privacy
As presented in Section IV, dummy keywords are inserted
into each data vector and some of them are selected in every
query. Therefore, similarity scores of documents will be not
exactly accurate. In other words, when cloud server returns
top-k documents based on similarity scores of data vectors to
query vector, some of real top-k relevant documents for the
query may be excluded. This is because either their original
similarity scores are decreased or the similarity scores of some
documents out of the real top-k are increased, both of which
are due to the impact of dummy keywords inserted into data
vectors. To evaluate the purity of the k documents retrieved
by user, we define a measure as precision Pk = k /k where k
is number of real top-k documents that are returned by cloud
server. Fig. 2(a) shows that the precision in MRSE III scheme
is evidently affected by the standard deviation σ of the random
variable ε. From the consideration of effectiveness, standard
deviation σ is expected to be smaller so as to obtain high
precision indicating the good purity of retrieved documents.
However, user privacy may have been partially leaked to
cloud server as a consequence of small σ. As described
in section III-B, access pattern is defined as the sequence
of ranked search results. Although search results cannot be
protected (excluding costly PIR technique), we can still hide
the rank order of retrieved documents as much as possible. In
order to evaluate this privacy guarantee, we first define the rank
perturbation as pi = |ri −r i|, where ri is the rank number of
document i in the retrieved top-k documents and r i is its rank
number in the real top-i ranked documents. The overall rank
8. 8
1 2 3 4 5 6 7
0.2
0.4
0.6
0.8
1
1.2
1.4
# of documents in dataset (x 10
2
)
Timeofgeneratingtradpoor(s)
MRSE_I
MRSE_II
MRSE_III
(a)
5 10 15 20 25 30 35
0.9
0.92
0.94
0.96
0.98
1
# of keywords in query
Timeofgeneratingtradpoor(s)
MRSE_I
MRSE_II
MRSE_III
(b)
Fig. 4: Time cost of generating trapdoor. (a) For the same
number (10) of keywords in a query within different number
of documents in dataset. (b) For different number of keywords
in a query within same number (500) of documents in dataset.
privacy measure at point k is then defined as the average of all
the pi for every document i in the retrieved top-k documents,
denoted as Pk = pi/k. Fig. 2(b) shows the rank privacy
at different points with two standard deviations σ = 1 and
σ = 0.5 respectively.
From these two figures, we can see that small σ leads to
higher precision of search result but lower rank privacy guaran-
tee, while large σ results in higher rank privacy guarantee but
lower precision. In other words, our scheme provides a balance
parameter for data users to satisfy their different requirements
on precision and rank privacy.
B. Efficiency
1) Index Construction: To build a searchable index I from
dataset F, the first step is to map the keyword set extracted
from each document Fi to a data vector Di, followed by
encrypting every data vector. The time cost of mapping or
encrypting depends directly on the dimensionality of data
vector which is determined by the number of distinct keywords
in the dataset. Fig. 3(a) shows that the number of documents in
dataset determines the number of distinct keywords. Note that
a list of standard IR techniques can be used to significantly
reduce the number of distinct keywords, such as case folding,
stemming, and stop words. We omit this refining process and
refer readers to [4] for more details. Fig. 3(b) shows that
the computation cost of building index is almost linear with
the number of documents in dataset. The index construction
computation cost in MRSE I scheme is very similar with that
in MRSE II scheme since the dimensionality in MRSE II is
just one more than that in the former scheme. The number
of dummy keywords in MRSE III scheme becomes 160 so
that the index construction time is slight larger than the
other two schemes. Although the time of building index is
not a negligible overhead for data owner, this is a one-time
operation before data outsourcing. Besides, Tab. V lists the
storage overhead of subindex in MRSE III scheme within
different number of documents in dataset. The subindex size
is absolutely linear with the dimensionality of data vector, but
its increasing speed slows down in coincidence with that of
the number of distinct keywords. The subindex size in the
1 2 3 4 5 6 7
0
20
40
60
80
# of documents in dataset (x 10
2
)
Timeofquery(ms)
MRSE_I
MRSE_II
MRSE_III
(a)
5 10 15 20 25 30 35
47
48
49
50
51
# of keywords in query
Timeofquery(ms)
MRSE_I
MRSE_II
MRSE_III
(b)
Fig. 5: Time cost of query. (a) For the same number (10) of
keywords in a query within different number of documents
in dataset. (b) For different number of keywords in a query
within the same number (500) of documents in dataset.
TABLE V: Storage overhead of subindex
# of documents in dataset 1000 2000 3000 4000 5000
Size of subindex (KB) 101.4 131.8 166.6 203.8 228.9
other two MRSE schemes is close to that in MRSE III scheme
because of trivial differences in dimensionality of data vector.
2) Trapdoor Generation: Fig. 4(a) shows that the time
to generate a trapdoor is greatly affected by the number of
documents in dataset. Like index construction, every trap-
door generation incurs two multiplications of a matrix and
a split query vector, where the dimensionality of matrix
or query vector is different in three proposed schemes and
becomes larger with the increasing number of documents in
dataset. Fig. 4(b) demonstrates the trapdoor generation cost in
MRSE III scheme is about 3 percentages larger than that in the
other two schemes, which is majorally brought by the larger
dimensionality. More importantly, it shows that the number
of keywords in a query have little influence on the overhead
of trapdoor generation, which is a significant advantage over
related works on multi-keyword searchable encryption.
3) Query: Query execution in cloud server consists of
computing and ranking similarity scores for all documents in
the dataset. Fig. 5 shows the query time is dominated by the
number of documents in dataset, and the number of keywords
in the query has very slight impact on it like the trapdoor
generation above. With respect to the communication cost in
Query, the size of trapdoor is the same as that of subindex
listed in the Tab. V, which keeps constant in the same dataset,
no matter how many keywords are contained in a query.
While the computation and communication cost in the query
procedure is linear with the number of query keywords in other
multiple-keyword search schemes [14], [16], our proposed
schemes enjoy the constant overhead in the query which makes
it more practical in the cloud paradigm.
VI. RELATED WORK
Single Keyword Searchable Encryption Traditional single
keyword searchable encryption schemes [5]–[13], [22] usually
build an encrypted searchable index such that its content is
hidden to the server unless it is given appropriate trapdoors
9. 9
generated via secret key(s) [2]. It is first studied by Song et
al. [5] in the symmetric key setting, and improvements and
advanced security definitions are given in Goh [6], Chang
et al. [7] and Curtmola et al. [8]. Our early work [22]
solves secure ranked keyword search which utilizes keyword
frequency to rank results instead of returning undifferentiated
results. However, it only supports single keyword search.
In the public key setting, Boneh et al. [9] present the first
searchable encryption construction, where anyone with public
key can write to the data stored on server but only authorized
users with private key can search. Public key solutions are
usually very computationally expensive however. Furthermore,
the keyword privacy could not be protected in the public key
setting since server could encrypt any keyword with public key
and then use the received trapdoor to evaluate this ciphertext.
Boolean Keyword Searchable Encryption To enrich search
functionalities, conjunctive keyword search [14]–[18] over
encrypted data have been proposed. These schemes incur
large overhead caused by their fundamental primitives, such as
computation cost by bilinear map, e.g. [16], or communication
cost by secret sharing, e.g. [15]. As a more general search
approach, predicate encryption schemes [19]–[21] are recently
proposed to support both conjunctive and disjunctive search.
Conjunctive keyword search returns “all-or-nothing”, which
means it only returns those documents in which all the
keywords specified by the search query appear; disjunctive
keyword search returns undifferentiated results, which means
it returns every document that contains a subset of the specific
keywords, even only one keyword of interest. In short, none
of existing Boolean keyword searchable encryption schemes
support multiple keywords ranked search over encrypted cloud
data while preserving privacy as we propose to explore in this
paper. Note that, inner product queries in predicate encryption
only predicates whether two vectors are orthogonal or not,
i.e., the inner product value is concealed except when it
equals zero. Without providing the capability to compare
concealed inner products, predicate encryption is not qualified
for performing ranked search. Furthermore, most of these
schemes are built upon the expensive evaluation of pairing
operations on elliptic curves. Such inefficiency disadvantage
also limits their practical performance when deployed in cloud.
On a different front, the research on top-k retrieval [24] in
database community is also loosely connected to our problem.
VII. CONCLUSION
In this paper, for the first time we define and solve the
problem of multi-keyword ranked search over encrypted cloud
data, and establish a variety of privacy requirements. Among
various multi-keyword semantics, we choose the efficient
principle of “coordinate matching”, i.e., as many matches
as possible, to effectively capture similarity between query
keywords and outsourced documents, and use “inner product
similarity” to quantitatively formalize such a principle for sim-
ilarity measurement. For meeting the challenge of supporting
multi-keyword semantic without privacy breaches, we first
propose a basic MRSE scheme using secure inner product
computation, and significantly improve it to achieve privacy
requirements in two levels of threat models. Thorough analysis
investigating privacy and efficiency guarantees of proposed
schemes is given, and experiments on the real-world dataset
show our proposed schemes introduce low overhead on both
computation and communication. As our future work, we
will explore supporting other multi-keyword semantics (e.g.,
weighted query) over encrypted data, integrity check of rank
order in search result and privacy guarantees in more stronger
threat model.
REFERENCES
[1] L. M. Vaquero, L. Rodero-Merino, J. Caceres, and M. Lindner, “A break
in the clouds: towards a cloud definition,” ACM SIGCOMM Comput.
Commun. Rev., vol. 39, no. 1, pp. 50–55, 2009.
[2] S. Kamara and K. Lauter, “Cryptographic cloud storage,” in RLCPS,
January 2010, LNCS. Springer, Heidelberg.
[3] A. Singhal, “Modern information retrieval: A brief overview,” IEEE
Data Engineering Bulletin, vol. 24, no. 4, pp. 35–43, 2001.
[4] I. H. Witten, A. Moffat, and T. C. Bell, “Managing gigabytes: Compress-
ing and indexing documents and images,” Morgan Kaufmann Publishing,
San Francisco, May 1999.
[5] D. Song, D. Wagner, and A. Perrig, “Practical techniques for searches
on encrypted data,” in Proc. of S&P, 2000.
[6] E.-J. Goh, “Secure indexes,” Cryptology ePrint Archive, 2003, http://
eprint.iacr.org/2003/216.
[7] Y.-C. Chang and M. Mitzenmacher, “Privacy preserving keyword
searches on remote encrypted data,” in Proc. of ACNS, 2005.
[8] R. Curtmola, J. A. Garay, S. Kamara, and R. Ostrovsky, “Searchable
symmetric encryption: improved definitions and efficient constructions,”
in Proc. of ACM CCS, 2006.
[9] D. Boneh, G. D. Crescenzo, R. Ostrovsky, and G. Persiano, “Public key
encryption with keyword search,” in Proc. of EUROCRYPT, 2004.
[10] M. Bellare, A. Boldyreva, and A. ONeill, “Deterministic and efficiently
searchable encryption,” in Proc. of CRYPTO, 2007.
[11] M. Abdalla, M. Bellare, D. Catalano, E. Kiltz, T. Kohno, T. Lange,
J. Malone-Lee, G. Neven, P. Paillier, and H. Shi, “Searchable encryp-
tion revisited: Consistency properties, relation to anonymous ibe, and
extensions,” J. Cryptol., vol. 21, no. 3, pp. 350–391, 2008.
[12] J. Li, Q. Wang, C. Wang, N. Cao, K. Ren, and W. Lou, “Fuzzy keyword
search over encrypted data in cloud computing,” in Proc. of IEEE
INFOCOM’10 Mini-Conference, San Diego, CA, USA, March 2010.
[13] D. Boneh, E. Kushilevitz, R. Ostrovsky, and W. E. S. III, “Public key
encryption that allows pir queries,” in Proc. of CRYPTO, 2007.
[14] P. Golle, J. Staddon, and B. Waters, “Secure conjunctive keyword search
over encrypted data,” in Proc. of ACNS, 2004, pp. 31–45.
[15] L. Ballard, S. Kamara, and F. Monrose, “Achieving efficient conjunctive
keyword searches over encrypted data,” in Proc. of ICICS, 2005.
[16] D. Boneh and B. Waters, “Conjunctive, subset, and range queries on
encrypted data,” in Proc. of TCC, 2007, pp. 535–554.
[17] R. Brinkman, “Searching in encrypted data,” in University of Twente,
PhD thesis, 2007.
[18] Y. Hwang and P. Lee, “Public key encryption with conjunctive keyword
search and its extension to a multi-user system,” in Pairing, 2007.
[19] J. Katz, A. Sahai, and B. Waters, “Predicate encryption supporting
disjunctions, polynomial equations, and inner products,” in Proc. of
EUROCRYPT, 2008.
[20] A. Lewko, T. Okamoto, A. Sahai, K. Takashima, and B. Waters,
“Fully secure functional encryption: Attribute-based encryption and
(hierarchical) inner product encryption,” in Proc. of EUROCRYPT, 2010.
[21] E. Shen, E. Shi, and B. Waters, “Predicate privacy in encryption
systems,” in Proc. of TCC, 2009.
[22] C. Wang, N. Cao, J. Li, K. Ren, and W. Lou, “Secure ranked keyword
search over encrypted cloud data,” in Proc. of ICDCS’10, 2010.
[23] S. Zerr, E. Demidova, D. Olmedilla, W. Nejdl, M. Winslett, and S. Mitra,
“Zerber: r-confidential indexing for distributed documents,” in Proc. of
EDBT, 2008, pp. 287–298.
[24] S. Zerr, D. Olmedilla, W. Nejdl, and W. Siberski, “Zerber+r: Top-k
retrieval from a confidential index,” in Proc. of EDBT, 2009.
[25] Y. Ishai, E. Kushilevitz, R. Ostrovsky, and A. Sahai, “Cryptography from
anonymity,” in Proc. of FOCS, 2006, pp. 239–248.
[26] W. K. Wong, D. W. Cheung, B. Kao, and N. Mamoulis, “Secure knn
computation on encrypted databases,” in Proc. of SIGMOD, 2009.
[27] W. W. Cohen, “Enron email dataset,” http://www.cs.cmu.edu/∼enron/.