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
Finding Relationships between the Our-NIR Cluster ResultsCSCJournals
The problem of evaluating node importance in clustering has been active research in present days and many methods have been developed. Most of the clustering algorithms deal with general similarity measures. However In real situation most of the cases data changes over time. But clustering this type of data not only decreases the quality of clusters but also disregards the expectation of users, when usually require recent clustering results. In this regard we proposed Our-NIR method that is better than Ming-Syan Chen proposed a method and it has proven with the help of results of node importance, which is related to calculate the node importance that is very useful in clustering of categorical data, still it has deficiency that is importance of data labeling and outlier detection. In this paper we modified Our-NIR method for evaluating of node importance by introducing the probability distribution which will be better than by comparing the results.
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology
AN EFFICIENT PARALLEL ALGORITHM FOR COMPUTING DETERMINANT OF NON-SQUARE MATRI...ijdpsjournal
One of the most significant challenges in Computing Determinant of Rectangular Matrices is high time
complexity of its algorithm. Among all definitions of determinant of rectangular matrices, Radic’s
definition has special features which make it more notable. But in this definition, C(N
M
) sub matrices of the
order m×m needed to be generated that put this problem in np-hard class. On the other hand, any row or
column reduction operation may hardly lead to diminish the volume of calculation. Therefore, in this paper
we try to present the parallel algorithm which can decrease the time complexity of computing the
determinant of non-square matrices to O(N).
Critical Paths Identification on Fuzzy Network Projectiosrjce
In this paper, a new approach for identifying fuzzy critical path is presented, based on converting the
fuzzy network project into deterministic network project, by transforming the parameters set of the fuzzy
activities into the time probability density function PDF of each fuzzy time activity. A case study is considered as
a numerical tested problem to demonstrate our approach.
Optimising Data Using K-Means Clustering AlgorithmIJERA Editor
K-means is one of the simplest unsupervised learning algorithms that solve the well known clustering problem. The procedure follows a simple and easy way to classify a given data set through a certain number of clusters (assume k clusters) fixed a priori. The main idea is to define k centroids, one for each cluster. These centroids should be placed in a cunning way because of different location causes different result. So, the better choice is to place them as much as possible far away from each other.
Finding Relationships between the Our-NIR Cluster ResultsCSCJournals
The problem of evaluating node importance in clustering has been active research in present days and many methods have been developed. Most of the clustering algorithms deal with general similarity measures. However In real situation most of the cases data changes over time. But clustering this type of data not only decreases the quality of clusters but also disregards the expectation of users, when usually require recent clustering results. In this regard we proposed Our-NIR method that is better than Ming-Syan Chen proposed a method and it has proven with the help of results of node importance, which is related to calculate the node importance that is very useful in clustering of categorical data, still it has deficiency that is importance of data labeling and outlier detection. In this paper we modified Our-NIR method for evaluating of node importance by introducing the probability distribution which will be better than by comparing the results.
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology
AN EFFICIENT PARALLEL ALGORITHM FOR COMPUTING DETERMINANT OF NON-SQUARE MATRI...ijdpsjournal
One of the most significant challenges in Computing Determinant of Rectangular Matrices is high time
complexity of its algorithm. Among all definitions of determinant of rectangular matrices, Radic’s
definition has special features which make it more notable. But in this definition, C(N
M
) sub matrices of the
order m×m needed to be generated that put this problem in np-hard class. On the other hand, any row or
column reduction operation may hardly lead to diminish the volume of calculation. Therefore, in this paper
we try to present the parallel algorithm which can decrease the time complexity of computing the
determinant of non-square matrices to O(N).
Critical Paths Identification on Fuzzy Network Projectiosrjce
In this paper, a new approach for identifying fuzzy critical path is presented, based on converting the
fuzzy network project into deterministic network project, by transforming the parameters set of the fuzzy
activities into the time probability density function PDF of each fuzzy time activity. A case study is considered as
a numerical tested problem to demonstrate our approach.
Optimising Data Using K-Means Clustering AlgorithmIJERA Editor
K-means is one of the simplest unsupervised learning algorithms that solve the well known clustering problem. The procedure follows a simple and easy way to classify a given data set through a certain number of clusters (assume k clusters) fixed a priori. The main idea is to define k centroids, one for each cluster. These centroids should be placed in a cunning way because of different location causes different result. So, the better choice is to place them as much as possible far away from each other.
Fault diagnosis using genetic algorithms and principal curveseSAT Journals
Abstract Several applications of nonlinear principal component analysis (NPCA) have appeared recently in process monitoring and fault diagnosis. In this paper a new approach is proposed for fault detection based on principal curves and genetic algorithms. The principal curve is a generation of linear principal component (PCA) introduced by Hastie as a parametric curve passes satisfactorily through the middle of data. The existing principal curves algorithms employ the first component of the data as an initial estimation of principal curve. However the dependence on initial line leads to a lack of flexibility and the final curve is only satisfactory for specific problems. In this paper we extend this work in two ways. First, we propose a new method based on genetic algorithms to find the principal curve. Here, lines are fitted and connected to form polygonal lines (PL). Second, potential application of principal curves is discussed. An example is used to illustrate fault diagnosis of nonlinear process using the proposed approach. Index Terms: Principal curve, Genetic Algorithm, Nonlinear principal component analysis, Fault detection.
International Journal of Engineering Research and Applications (IJERA) is an open access online peer reviewed international journal that publishes research and review articles in the fields of Computer Science, Neural Networks, Electrical Engineering, Software Engineering, Information Technology, Mechanical Engineering, Chemical Engineering, Plastic Engineering, Food Technology, Textile Engineering, Nano Technology & science, Power Electronics, Electronics & Communication Engineering, Computational mathematics, Image processing, Civil Engineering, Structural Engineering, Environmental Engineering, VLSI Testing & Low Power VLSI Design etc.
STUDY ANALYSIS ON TRACKING MULTIPLE OBJECTS IN PRESENCE OF INTER OCCLUSION IN...aciijournal
The object tracking algorithm is used to tracking multiple objects in a video streams. This paper provides
Mutual tracking algorithm which improve the estimation inaccuracy and the robustness of clutter
environment when it uses Kalman Filter. using this algorithm to avoid the problem of id switch in
continuing occlusions. First the algorithms apply the collision avoidance model to separate the nearby
trajectories. Suppose occurring inter occlusion the aggregate model splits into several parts and use only
visible parts perform tracking. The algorithm reinitializes the particles when the tracker is fully occluded.
The experimental results using unmanned level crossing (LC) exhibit the feasibility of our proposal. In
addition, comparison with Kalman filter trackers has also been performed.
Dimensionality Reduction Techniques for Document Clustering- A SurveyIJTET Journal
Abstract— Dimensionality reduction technique is applied to get rid of the inessential terms like redundant and noisy terms in documents. In this paper a systematic study is conducted for seven dimensionality reduction methods such as Latent Semantic Indexing (LSI), Random Projection (RP), Principle Component Analysis (PCA) and CUR decomposition, Latent Dirichlet Allocation(LDA), Singular value decomposition (SVD). Linear Discriminant Analysis(LDA)
SVD BASED LATENT SEMANTIC INDEXING WITH USE OF THE GPU COMPUTATIONSijscmcj
The purpose of this article is to determine the usefulness of the Graphics Processing Unit (GPU) calculations used to implement the Latent Semantic Indexing (LSI) reduction of the TERM-BY DOCUMENT matrix. Considered reduction of the matrix is based on the use of the SVD (Singular Value Decomposition) decomposition. A high computational complexity of the SVD decomposition - O(n3), causes that a reduction of a large indexing structure is a difficult task. In this article there is a comparison of the time complexity and accuracy of the algorithms implemented for two different environments. The first environment is associated with the CPU and MATLAB R2011a. The second environment is related to graphics processors and the CULA library. The calculations were carried out on generally available benchmark matrices, which were combined to achieve the resulting matrix of high size. For both considered environments computations were performed for double and single precision data.
Anomaly Detection in Temporal data Using Kmeans Clustering with C5.0theijes
Anomaly detection is a challenging problem in Temporal data .In this paper we have proposed an algorithm using two different machine learning techniques Kmeans clustering and C5.0 decision tree , where Euclidean distance is used to find the closest cluster for the data set and then decision tree is built for each cluster using C5.0 decision tree technique and the rules of decision tree is used to classify each anomalous and normal instances in the dataset .The proposed algorithm gives impressive classification accuracy in the experimented result and describe the proposed system of kmeans and C5.0 decision tree
A Counterexample to the Forward Recursion in Fuzzy Critical Path Analysis Und...ijfls
Fuzzy logic is an alternate approach for quantifying uncertainty relating to activity duration. The fuzzy
version of the backward recursion has been shown to produce results that incorrectly amplify the level of
uncertainty. However, the fuzzy version of the forward recursion has been widely proposed as an
approach for determining the fuzzy set of critical path lengths. In this paper, the direct application of the
extension principle leads to a proposition that must be satisfied in fuzzy critical path analysis. Using a
counterexample it is demonstrated that the fuzzy forward recursion when discrete fuzzy sets are used to
represent activity durations produces results that are not consistent with the theory presented. The
problem is shown to be the application of the fuzzy maximum. Several methods presented in the literature
are described and shown to provide results that are consistent with the extension principle.
A PSO-Based Subtractive Data Clustering AlgorithmIJORCS
There is a tremendous proliferation in the amount of information available on the largest shared information source, the World Wide Web. Fast and high-quality clustering algorithms play an important role in helping users to effectively navigate, summarize, and organize the information. Recent studies have shown that partitional clustering algorithms such as the k-means algorithm are the most popular algorithms for clustering large datasets. The major problem with partitional clustering algorithms is that they are sensitive to the selection of the initial partitions and are prone to premature converge to local optima. Subtractive clustering is a fast, one-pass algorithm for estimating the number of clusters and cluster centers for any given set of data. The cluster estimates can be used to initialize iterative optimization-based clustering methods and model identification methods. In this paper, we present a hybrid Particle Swarm Optimization, Subtractive + (PSO) clustering algorithm that performs fast clustering. For comparison purpose, we applied the Subtractive + (PSO) clustering algorithm, PSO, and the Subtractive clustering algorithms on three different datasets. The results illustrate that the Subtractive + (PSO) clustering algorithm can generate the most compact clustering results as compared to other algorithms.
New approach for wolfe’s modified simplex method to solve quadratic programmi...eSAT Journals
Abstract
In this paper, an alternative method for Wolfe’s modified simplex method is introduced. This method is easy to solve quadratic programming problem (QPP) concern with non-linear programming problem (NLPP). In linear programming models, the characteristic assumption is the linearity of the objective function and constraints. Although this assumption holds in numerous practical situations, yet we come across many situations where the objective function and some or all of the constraints are non-linear functions. The non-linearity of the functions makes the solution of the problem much more involved as compared to LPPs and there is no single algorithm like the simplex method, which can be employed to solve efficiently all NPPs.
Keywords: Quadratic programming problem, New approach, Modified simplex method, and Optimal solution.
In this paper generation of binary sequences derived from chaotic sequences defined over Z4 is proposed.
The six chaotic map equations considered in this paper are Logistic map, Tent Map, Cubic Map, Quadratic
Map and Bernoulli Map. Using these chaotic map equations, sequences over Z4 are generated which are
converted to binary sequences using polynomial mapping. Segments of sequences of different lengths are
tested for cross correlation and linear complexity properties. It is found that some segments of different
length of these sequences have good cross correlation and linear complexity properties. The Bit Error Rate
performance in DS-CDMA communication systems using these binary sequences is found to be better than
Gold sequences and Kasami sequences.
Privacy Policy Inference of User-Uploaded Images on Content Sharing Sites1crore 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
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
Fault diagnosis using genetic algorithms and principal curveseSAT Journals
Abstract Several applications of nonlinear principal component analysis (NPCA) have appeared recently in process monitoring and fault diagnosis. In this paper a new approach is proposed for fault detection based on principal curves and genetic algorithms. The principal curve is a generation of linear principal component (PCA) introduced by Hastie as a parametric curve passes satisfactorily through the middle of data. The existing principal curves algorithms employ the first component of the data as an initial estimation of principal curve. However the dependence on initial line leads to a lack of flexibility and the final curve is only satisfactory for specific problems. In this paper we extend this work in two ways. First, we propose a new method based on genetic algorithms to find the principal curve. Here, lines are fitted and connected to form polygonal lines (PL). Second, potential application of principal curves is discussed. An example is used to illustrate fault diagnosis of nonlinear process using the proposed approach. Index Terms: Principal curve, Genetic Algorithm, Nonlinear principal component analysis, Fault detection.
International Journal of Engineering Research and Applications (IJERA) is an open access online peer reviewed international journal that publishes research and review articles in the fields of Computer Science, Neural Networks, Electrical Engineering, Software Engineering, Information Technology, Mechanical Engineering, Chemical Engineering, Plastic Engineering, Food Technology, Textile Engineering, Nano Technology & science, Power Electronics, Electronics & Communication Engineering, Computational mathematics, Image processing, Civil Engineering, Structural Engineering, Environmental Engineering, VLSI Testing & Low Power VLSI Design etc.
STUDY ANALYSIS ON TRACKING MULTIPLE OBJECTS IN PRESENCE OF INTER OCCLUSION IN...aciijournal
The object tracking algorithm is used to tracking multiple objects in a video streams. This paper provides
Mutual tracking algorithm which improve the estimation inaccuracy and the robustness of clutter
environment when it uses Kalman Filter. using this algorithm to avoid the problem of id switch in
continuing occlusions. First the algorithms apply the collision avoidance model to separate the nearby
trajectories. Suppose occurring inter occlusion the aggregate model splits into several parts and use only
visible parts perform tracking. The algorithm reinitializes the particles when the tracker is fully occluded.
The experimental results using unmanned level crossing (LC) exhibit the feasibility of our proposal. In
addition, comparison with Kalman filter trackers has also been performed.
Dimensionality Reduction Techniques for Document Clustering- A SurveyIJTET Journal
Abstract— Dimensionality reduction technique is applied to get rid of the inessential terms like redundant and noisy terms in documents. In this paper a systematic study is conducted for seven dimensionality reduction methods such as Latent Semantic Indexing (LSI), Random Projection (RP), Principle Component Analysis (PCA) and CUR decomposition, Latent Dirichlet Allocation(LDA), Singular value decomposition (SVD). Linear Discriminant Analysis(LDA)
SVD BASED LATENT SEMANTIC INDEXING WITH USE OF THE GPU COMPUTATIONSijscmcj
The purpose of this article is to determine the usefulness of the Graphics Processing Unit (GPU) calculations used to implement the Latent Semantic Indexing (LSI) reduction of the TERM-BY DOCUMENT matrix. Considered reduction of the matrix is based on the use of the SVD (Singular Value Decomposition) decomposition. A high computational complexity of the SVD decomposition - O(n3), causes that a reduction of a large indexing structure is a difficult task. In this article there is a comparison of the time complexity and accuracy of the algorithms implemented for two different environments. The first environment is associated with the CPU and MATLAB R2011a. The second environment is related to graphics processors and the CULA library. The calculations were carried out on generally available benchmark matrices, which were combined to achieve the resulting matrix of high size. For both considered environments computations were performed for double and single precision data.
Anomaly Detection in Temporal data Using Kmeans Clustering with C5.0theijes
Anomaly detection is a challenging problem in Temporal data .In this paper we have proposed an algorithm using two different machine learning techniques Kmeans clustering and C5.0 decision tree , where Euclidean distance is used to find the closest cluster for the data set and then decision tree is built for each cluster using C5.0 decision tree technique and the rules of decision tree is used to classify each anomalous and normal instances in the dataset .The proposed algorithm gives impressive classification accuracy in the experimented result and describe the proposed system of kmeans and C5.0 decision tree
A Counterexample to the Forward Recursion in Fuzzy Critical Path Analysis Und...ijfls
Fuzzy logic is an alternate approach for quantifying uncertainty relating to activity duration. The fuzzy
version of the backward recursion has been shown to produce results that incorrectly amplify the level of
uncertainty. However, the fuzzy version of the forward recursion has been widely proposed as an
approach for determining the fuzzy set of critical path lengths. In this paper, the direct application of the
extension principle leads to a proposition that must be satisfied in fuzzy critical path analysis. Using a
counterexample it is demonstrated that the fuzzy forward recursion when discrete fuzzy sets are used to
represent activity durations produces results that are not consistent with the theory presented. The
problem is shown to be the application of the fuzzy maximum. Several methods presented in the literature
are described and shown to provide results that are consistent with the extension principle.
A PSO-Based Subtractive Data Clustering AlgorithmIJORCS
There is a tremendous proliferation in the amount of information available on the largest shared information source, the World Wide Web. Fast and high-quality clustering algorithms play an important role in helping users to effectively navigate, summarize, and organize the information. Recent studies have shown that partitional clustering algorithms such as the k-means algorithm are the most popular algorithms for clustering large datasets. The major problem with partitional clustering algorithms is that they are sensitive to the selection of the initial partitions and are prone to premature converge to local optima. Subtractive clustering is a fast, one-pass algorithm for estimating the number of clusters and cluster centers for any given set of data. The cluster estimates can be used to initialize iterative optimization-based clustering methods and model identification methods. In this paper, we present a hybrid Particle Swarm Optimization, Subtractive + (PSO) clustering algorithm that performs fast clustering. For comparison purpose, we applied the Subtractive + (PSO) clustering algorithm, PSO, and the Subtractive clustering algorithms on three different datasets. The results illustrate that the Subtractive + (PSO) clustering algorithm can generate the most compact clustering results as compared to other algorithms.
New approach for wolfe’s modified simplex method to solve quadratic programmi...eSAT Journals
Abstract
In this paper, an alternative method for Wolfe’s modified simplex method is introduced. This method is easy to solve quadratic programming problem (QPP) concern with non-linear programming problem (NLPP). In linear programming models, the characteristic assumption is the linearity of the objective function and constraints. Although this assumption holds in numerous practical situations, yet we come across many situations where the objective function and some or all of the constraints are non-linear functions. The non-linearity of the functions makes the solution of the problem much more involved as compared to LPPs and there is no single algorithm like the simplex method, which can be employed to solve efficiently all NPPs.
Keywords: Quadratic programming problem, New approach, Modified simplex method, and Optimal solution.
In this paper generation of binary sequences derived from chaotic sequences defined over Z4 is proposed.
The six chaotic map equations considered in this paper are Logistic map, Tent Map, Cubic Map, Quadratic
Map and Bernoulli Map. Using these chaotic map equations, sequences over Z4 are generated which are
converted to binary sequences using polynomial mapping. Segments of sequences of different lengths are
tested for cross correlation and linear complexity properties. It is found that some segments of different
length of these sequences have good cross correlation and linear complexity properties. The Bit Error Rate
performance in DS-CDMA communication systems using these binary sequences is found to be better than
Gold sequences and Kasami sequences.
Privacy Policy Inference of User-Uploaded Images on Content Sharing Sites1crore 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
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
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
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
Context-Based Diversification for Keyword Queries over XML Data1crore 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
Reverse Nearest Neighbors in Unsupervised Distance-Based Outlier Detection1crore 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
On Summarization and Timeline Generation for Evolutionary Tweet Streams1crore 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
The well-known saying is that a picture is worth a thousand words; but what if you’re the picture, and you’re being shown off to a gorgeous woman, or even perhaps a prospective employer? Would you like the way you look? Are you happy with the vibes that you think you give out? What are your eyes, hands and shoulders saying? A little worried aren’t we?
A Novel Neuroglial Architecture for Modelling Singular Perturbation System IJECEIAES
This work develops a new modular architecture that emulates a recentlydiscovered biological paradigm. It originates from the human brain where the information flows along two different pathways and is processed along two time scales: one is a fast neural network (NN) and the other is a slow network called the glial network (GN). It was found that the neural network is powered and controlled by the glial network. Based on our biological knowledge of glial cells and the powerful concept of modularity, a novel approach called artificial neuroglial Network (ANGN) was designed and an algorithm based on different concepts of modularity was also developed. The implementation is based on the notion of multi-time scale systems. Validation is performed through an asynchronous machine (ASM) modeled in the standard singularly perturbed form. We apply the geometrical approach, based on Gerschgorin’s circle theorem (GCT), to separate the fast and slow variables, as well as the singular perturbation method (SPM) to determine the reduced models. This new architecture makes it possible to obtain smaller networks with less complexity and better performance.
DCE: A NOVEL DELAY CORRELATION MEASUREMENT FOR TOMOGRAPHY WITH PASSIVE REAL...ijdpsjournal
Tomography is important for network design and routing optimization. Prior approaches require either
precise time synchronization or complex cooperation. Furthermore, active tomography consumes explicit
probing resulting in limited scalability. To address the first issue we propose a novel Delay Correlation
Estimation methodology named DCE with no need of synchronization and special cooperation. For the
second issue we develop a passive realization mechanism merely using regular data flow without explicit
bandwidth consumption. Extensive simulations in OMNeT++ are made to evaluate its accuracy where we
show that DCE measurement is highly identical with the true value. Also from test result we find that
mechanism of passive realization is able to achieve both regular data transmission and purpose of
tomography with excellent robustness versus different background traffic and package size.
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology.
Implementation of low power divider techniques using radixeSAT Journals
Abstract
This work describes the design of a divder technique Low-power techniques are applied in the design of the unit, and energy-delay
tradeoffs considered. The energy dissipation in the divider can be reduced by up to 70% with respect to a standard implementation not
optimized for energy, without penalizing the latency. In this dividing technique we compare the radix-8 divider is compared with one
obtained by overlapping three radix-2 stages and with a radix-4 divider. Results show that the latency of our divider is similar to that
of the divider with overlapped stages, but the area is smaller. The speed-up of the radix-8 over the radix-4 is about 23.58% and the
energy dissipated to complete a division is almost the same, although the area of the radix-8 is 67.58% larger.
COMPARATIVE PERFORMANCE ANALYSIS OF RNSC AND MCL ALGORITHMS ON POWER-LAW DIST...acijjournal
Cluster analysis of graph related problems is an important issue now-a-day. Different types of graph
clustering techniques are appeared in the field but most of them are vulnerable in terms of effectiveness
and fragmentation of output in case of real-world applications in diverse systems. In this paper, we will
provide a comparative behavioural analysis of RNSC (Restricted Neighbourhood Search Clustering) and
MCL (Markov Clustering) algorithms on Power-Law Distribution graphs. RNSC is a graph clustering
technique using stochastic local search. RNSC algorithm tries to achieve optimal cost clustering by
assigning some cost functions to the set of clusterings of a graph. This algorithm was implemented by A.
D. King only for undirected and unweighted random graphs. Another popular graph clustering
algorithm MCL is based on stochastic flow simulation model for weighted graphs. There are plentiful
applications of power-law or scale-free graphs in nature and society. Scale-free topology is stochastic i.e.
nodes are connected in a random manner. Complex network topologies like World Wide Web, the web of
human sexual contacts, or the chemical network of a cell etc., are basically following power-law
distribution to represent different real-life systems. This paper uses real large-scale power-law
distribution graphs to conduct the performance analysis of RNSC behaviour compared with Markov
clustering (MCL) algorithm. Extensive experimental results on several synthetic and real power-law
distribution datasets reveal the effectiveness of our approach to comparative performance measure of
these algorithms on the basis of cost of clustering, cluster size, modularity index of clustering results and
normalized mutual information (NMI).
Time of arrival based localization in wireless sensor networks a non linear ...sipij
In this paper, we aim to obtain the location information of a sensor node deployed in a Wireless Sensor Network (WSN). Here, Time of Arrival based localization technique is considered. We calculate the position information of an unknown sensor node using the non- linear techniques. The performances of the techniques are compared with the Cramer Rao Lower bound (CRLB). Non-linear Least Squares and the Maximum Likelihood are the non-linear techniques that have been used to estimate the position of the unknown sensor node. Each of these non-linear techniques are iterative approaches, namely, Newton
Raphson estimate, Gauss Newton Estimate and the Steepest Descent estimate for comparison. Based on the
results of the simulation, the approaches have been compared. From the simulation study, Localization
based on Maximum Likelihood approach is having higher localization accuracy.
Design and Implementation of Variable Radius Sphere Decoding Algorithmcsandit
Sphere Decoding (SD) algorithm is an implement deco
ding algorithm based on Zero Forcing
(ZF) algorithm in the real number field. The classi
cal SD algorithm is famous for its
outstanding Bit Error Rate (BER) performance and de
coding strategy. The algorithm gets its
maximum likelihood solution by recursive shrinking
the searching radius gradually. However, it
is too complicated to use the method of shrinking t
he searching radius in ground
communication system. This paper proposed a Variabl
e Radius Sphere Decoding (VR-SD)
algorithm based on ZF algorithm in order to simplif
y the complex searching steps. We prove the
advantages of VR-SD algorithm by analyzing from the
derivation of mathematical formulas and
the simulation of the BER performance between SD an
d VR-SD algorithm.
NONLINEAR MODELING AND ANALYSIS OF WSN NODE LOCALIZATION METHODijwmn
In this paper, node localization algorithms in wireless sensor networks are researched, the traditional algorithms are studied, and some meaningful results are obtained. For the localization algorithm and route planning of WSN exists a big localization error in wireless communication. WSN communication system is researched. According to the anchor nodes and unknown nodes, a new localization algorithm and route planning method of WSN are proposed in this paper. At the same time, a new genetic algorithm of route planning of WSN is proposed. The performance of the node density and localization error is simulated and analyzed. The simulation results show that the performance of proposed WSN localization algorithm and route planning method are better than the traditional algorithms.
NONLINEAR MODELING AND ANALYSIS OF WSN NODE LOCALIZATION METHODijwmn
In this paper, node localization algorithms in wireless sensor networks are researched, the traditional
algorithms are studied, and some meaningful results are obtained. For the localization algorithm and route
planning of WSN exists a big localization error in wireless communication. WSN communication system is
researched. According to the anchor nodes and unknown nodes, a new localization algorithm and route
planning method of WSN are proposed in this paper. At the same time, a new genetic algorithm of route
planning of WSN is proposed. The performance of the node density and localization error is simulated and
analyzed. The simulation results show that the performance of proposed WSN localization algorithm and
route planning method are better than the traditional algorithms.
Tomography is important for network design and routing optimization. Prior approaches require either
precise time synchronization or complex cooperation. Furthermore, active tomography consumes explicit
probing resulting in limited scalability. To address the first issue we propose a novel Delay Correlation
Estimation methodology named DCE with no need of synchronization and special cooperation. For the
second issue we develop a passive realization mechanism merely using regular data flow without explicit
bandwidth consumption. Extensive simulations in OMNeT++ are made to evaluate its accuracy where we
show that DCE measurement is highly identical with the true value. Also from test result we find that
mechanism of passive realization is able to achieve both regular data transmission and purpose of
tomography with excellent robustness versus different background traffic and package size.
Performance Improvement of Vector Quantization with Bit-parallelism HardwareCSCJournals
Vector quantization is an elementary technique for image compression; however, searching for the nearest codeword in a codebook is time-consuming. In this work, we propose a hardware-based scheme by adopting bit-parallelism to prune unnecessary codewords. The new scheme uses a “Bit-mapped Look-up Table” to represent the positional information of the codewords. The lookup procedure can simply refer to the bitmaps to find the candidate codewords. Our simulation results further confirm the effectiveness of the proposed scheme.
MARGINAL PERCEPTRON FOR NON-LINEAR AND MULTI CLASS CLASSIFICATION ijscai
Generalization error of classifier can be reduced by larger margin of separating hyperplane. The proposed classification algorithm implements margin in classical perceptron algorithm, to reduce generalized errors by maximizing margin of separating hyperplane. Algorithm uses the same updation rule with the perceptron, to converge in a finite number of updates to solutions, possessing any desirable fraction of the margin. This solution is again optimized to get maximum possible margin. The algorithm can process linear, non-linear and multi class problems. Experimental results place the proposed classifier equivalent to the support vector machine and even better in some cases. Some preliminary experimental results are briefly discussed.
Multimode system condition monitoring using sparsity reconstruction for quali...IJECEIAES
In this paper, we introduce an improved multivariate statistical monitoring method based on the stacked sparse autoencoder (SSAE). Our contribution focuses on the choice of the SSAE model based on neural networks to solve diagnostic problems of complex systems. In order to monitor the process performance, the squared prediction error (SPE) chart is linked with nonparametric adaptive confidence bounds which arise from the kernel density estimation to minimize erroneous alerts. Then, faults are localized using two methods: contribution plots and sensor validity index (SVI). The results are obtained from experiments and real data from a drinkable water processing plant, demonstrating how the applied technique is performed. The simulation results of the SSAE model show a better ability to detect and identify sensor failures.
COMPARISON OF VOLUME AND DISTANCE CONSTRAINT ON HYPERSPECTRAL UNMIXINGcsandit
Algorithms based on minimum volume constraint or sum of squared distances constraint is
widely used in Hyperspectral image unmixing. However, there are few works about performing
comparison between these two algorithms. In this paper, comparison analysis between two
algorithms is presented to evaluate the performance of two constraints under different situations. Comparison is implemented from the following three aspects: flatness of simplex, initialization effects and robustness to noise. The analysis can provide a guideline on which constraint should be adopted under certain specific tasks.
Parallel Batch-Dynamic Graphs: Algorithms and Lower BoundsSubhajit Sahu
Highlighted notes on Parallel Batch-Dynamic Graphs: Algorithms and Lower Bounds.
While doing research work under Prof. Kishore Kothapalli.
Laxman Dhulipala, David Durfee, Janardhan Kulkarni, Richard Peng, Saurabh Sawlani, Xiaorui Sun:
Parallel Batch-Dynamic Graphs: Algorithms and Lower Bounds. SODA 2020: 1300-1319
In this paper we study the problem of dynamically maintaining graph properties under batches of edge insertions and deletions in the massively parallel model of computation. In this setting, the graph is stored on a number of machines, each having space strongly sublinear with respect to the number of vertices, that is, n for some constant 0 < < 1. Our goal is to handle batches of updates and queries where the data for each batch fits onto one machine in constant rounds of parallel computation, as well as to reduce the total communication between the machines. This objective corresponds to the gradual buildup of databases over time, while the goal of obtaining constant rounds of communication for problems in the static setting has been elusive for problems as simple as undirected graph connectivity. We give an algorithm for dynamic graph connectivity in this setting with constant communication rounds and communication cost almost linear in terms of the batch size. Our techniques combine a new graph contraction technique, an independent random sample extractor from correlated samples, as well as distributed data structures supporting parallel updates and queries in batches. We also illustrate the power of dynamic algorithms in the MPC model by showing that the batched version of the adaptive connectivity problem is P-complete in the centralized setting, but sub-linear sized batches can be handled in a constant number of rounds. Due to the wide applicability of our approaches, we believe it represents a practically-motivated workaround to the current difficulties in designing more efficient massively parallel static graph algorithms.
Similar to Scalable Constrained Spectral Clustering (20)
Unit 8 - Information and Communication Technology (Paper I).pdfThiyagu K
This slides describes the basic concepts of ICT, basics of Email, Emerging Technology and Digital Initiatives in Education. This presentations aligns with the UGC Paper I syllabus.
Synthetic Fiber Construction in lab .pptxPavel ( NSTU)
Synthetic fiber production is a fascinating and complex field that blends chemistry, engineering, and environmental science. By understanding these aspects, students can gain a comprehensive view of synthetic fiber production, its impact on society and the environment, and the potential for future innovations. Synthetic fibers play a crucial role in modern society, impacting various aspects of daily life, industry, and the environment. ynthetic fibers are integral to modern life, offering a range of benefits from cost-effectiveness and versatility to innovative applications and performance characteristics. While they pose environmental challenges, ongoing research and development aim to create more sustainable and eco-friendly alternatives. Understanding the importance of synthetic fibers helps in appreciating their role in the economy, industry, and daily life, while also emphasizing the need for sustainable practices and innovation.
Biological screening of herbal drugs: Introduction and Need for
Phyto-Pharmacological Screening, New Strategies for evaluating
Natural Products, In vitro evaluation techniques for Antioxidants, Antimicrobial and Anticancer drugs. In vivo evaluation techniques
for Anti-inflammatory, Antiulcer, Anticancer, Wound healing, Antidiabetic, Hepatoprotective, Cardio protective, Diuretics and
Antifertility, Toxicity studies as per OECD guidelines
The Roman Empire A Historical Colossus.pdfkaushalkr1407
The Roman Empire, a vast and enduring power, stands as one of history's most remarkable civilizations, leaving an indelible imprint on the world. It emerged from the Roman Republic, transitioning into an imperial powerhouse under the leadership of Augustus Caesar in 27 BCE. This transformation marked the beginning of an era defined by unprecedented territorial expansion, architectural marvels, and profound cultural influence.
The empire's roots lie in the city of Rome, founded, according to legend, by Romulus in 753 BCE. Over centuries, Rome evolved from a small settlement to a formidable republic, characterized by a complex political system with elected officials and checks on power. However, internal strife, class conflicts, and military ambitions paved the way for the end of the Republic. Julius Caesar’s dictatorship and subsequent assassination in 44 BCE created a power vacuum, leading to a civil war. Octavian, later Augustus, emerged victorious, heralding the Roman Empire’s birth.
Under Augustus, the empire experienced the Pax Romana, a 200-year period of relative peace and stability. Augustus reformed the military, established efficient administrative systems, and initiated grand construction projects. The empire's borders expanded, encompassing territories from Britain to Egypt and from Spain to the Euphrates. Roman legions, renowned for their discipline and engineering prowess, secured and maintained these vast territories, building roads, fortifications, and cities that facilitated control and integration.
The Roman Empire’s society was hierarchical, with a rigid class system. At the top were the patricians, wealthy elites who held significant political power. Below them were the plebeians, free citizens with limited political influence, and the vast numbers of slaves who formed the backbone of the economy. The family unit was central, governed by the paterfamilias, the male head who held absolute authority.
Culturally, the Romans were eclectic, absorbing and adapting elements from the civilizations they encountered, particularly the Greeks. Roman art, literature, and philosophy reflected this synthesis, creating a rich cultural tapestry. Latin, the Roman language, became the lingua franca of the Western world, influencing numerous modern languages.
Roman architecture and engineering achievements were monumental. They perfected the arch, vault, and dome, constructing enduring structures like the Colosseum, Pantheon, and aqueducts. These engineering marvels not only showcased Roman ingenuity but also served practical purposes, from public entertainment to water supply.
June 3, 2024 Anti-Semitism Letter Sent to MIT President Kornbluth and MIT Cor...Levi Shapiro
Letter from the Congress of the United States regarding Anti-Semitism sent June 3rd to MIT President Sally Kornbluth, MIT Corp Chair, Mark Gorenberg
Dear Dr. Kornbluth and Mr. Gorenberg,
The US House of Representatives is deeply concerned by ongoing and pervasive acts of antisemitic
harassment and intimidation at the Massachusetts Institute of Technology (MIT). Failing to act decisively to ensure a safe learning environment for all students would be a grave dereliction of your responsibilities as President of MIT and Chair of the MIT Corporation.
This Congress will not stand idly by and allow an environment hostile to Jewish students to persist. The House believes that your institution is in violation of Title VI of the Civil Rights Act, and the inability or
unwillingness to rectify this violation through action requires accountability.
Postsecondary education is a unique opportunity for students to learn and have their ideas and beliefs challenged. However, universities receiving hundreds of millions of federal funds annually have denied
students that opportunity and have been hijacked to become venues for the promotion of terrorism, antisemitic harassment and intimidation, unlawful encampments, and in some cases, assaults and riots.
The House of Representatives will not countenance the use of federal funds to indoctrinate students into hateful, antisemitic, anti-American supporters of terrorism. Investigations into campus antisemitism by the Committee on Education and the Workforce and the Committee on Ways and Means have been expanded into a Congress-wide probe across all relevant jurisdictions to address this national crisis. The undersigned Committees will conduct oversight into the use of federal funds at MIT and its learning environment under authorities granted to each Committee.
• The Committee on Education and the Workforce has been investigating your institution since December 7, 2023. The Committee has broad jurisdiction over postsecondary education, including its compliance with Title VI of the Civil Rights Act, campus safety concerns over disruptions to the learning environment, and the awarding of federal student aid under the Higher Education Act.
• The Committee on Oversight and Accountability is investigating the sources of funding and other support flowing to groups espousing pro-Hamas propaganda and engaged in antisemitic harassment and intimidation of students. The Committee on Oversight and Accountability is the principal oversight committee of the US House of Representatives and has broad authority to investigate “any matter” at “any time” under House Rule X.
• The Committee on Ways and Means has been investigating several universities since November 15, 2023, when the Committee held a hearing entitled From Ivory Towers to Dark Corners: Investigating the Nexus Between Antisemitism, Tax-Exempt Universities, and Terror Financing. The Committee followed the hearing with letters to those institutions on January 10, 202
Embracing GenAI - A Strategic ImperativePeter Windle
Artificial Intelligence (AI) technologies such as Generative AI, Image Generators and Large Language Models have had a dramatic impact on teaching, learning and assessment over the past 18 months. The most immediate threat AI posed was to Academic Integrity with Higher Education Institutes (HEIs) focusing their efforts on combating the use of GenAI in assessment. Guidelines were developed for staff and students, policies put in place too. Innovative educators have forged paths in the use of Generative AI for teaching, learning and assessments leading to pockets of transformation springing up across HEIs, often with little or no top-down guidance, support or direction.
This Gasta posits a strategic approach to integrating AI into HEIs to prepare staff, students and the curriculum for an evolving world and workplace. We will highlight the advantages of working with these technologies beyond the realm of teaching, learning and assessment by considering prompt engineering skills, industry impact, curriculum changes, and the need for staff upskilling. In contrast, not engaging strategically with Generative AI poses risks, including falling behind peers, missed opportunities and failing to ensure our graduates remain employable. The rapid evolution of AI technologies necessitates a proactive and strategic approach if we are to remain relevant.
How to Make a Field invisible in Odoo 17Celine George
It is possible to hide or invisible some fields in odoo. Commonly using “invisible” attribute in the field definition to invisible the fields. This slide will show how to make a field invisible in odoo 17.
1. Scalable Constrained Spectral Clustering
Jianyuan Li, Member, IEEE,
Yingjie Xia, Member, IEEE, Zhenyu Shan, and
Yuncai Liu, Senior Member, IEEE
Abstract—Constrained spectral clustering (CSC) algorithms have shown great
promise in significantly improving clustering accuracy by encoding side information
into spectral clustering algorithms. However, existing CSC algorithms are ineffi-
cient in handling moderate and large datasets. In this paper, we aim to develop a
scalable and efficient CSC algorithm by integrating sparse coding based graph
construction into a framework called constrained normalized cuts. To this end, we
formulate a scalable constrained normalized-cuts problem and solve it based on a
closed-form mathematical analysis. We demonstrate that this problem can be
reduced to a generalized eigenvalue problem that can be solved very efficiently.
We also describe a principled k-way CSC algorithm for handling moderate and
large datasets. Experimental results over benchmark datasets demonstrate that
the proposed algorithm is greatly cost-effective, in the sense that (1) with less side
information, it can obtain significant improvements in accuracy compared to the
unsupervised baseline; (2) with less computational time, it can achieve high clus-
tering accuracies close to those of the state-of-the-art.
Index Terms—Constrained spectral clustering, sparse coding, efficiency,
scalability
Ç
1 INTRODUCTION
CURRENTLY, data in a wide variety of areas tend to large scales. For
many traditional learning based data mining algorithms, it is a big
challenge to efficiently mine knowledge from the fast increasing
data such as information streams, images and even videos. To over-
come the challenge, it is important to develop scalable learning
algorithms.
Constrained clustering is an important area in the research
communities of machine learning. Researchers proposed many
new algorithms [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11],
[12], [13], [14], [15], [16], which improve clustering accuracy by
means of encoding side information into unsupervised clustering
algorithms. Here, side information might be labelled data [15],
pairwise constraints [2], relative comparison constraints [17],
and so forth. In this paper, we refer to it as labelled data. In
practice, labelled data are often costly to obtain and so the typi-
cal problem in this area is to improve clustering by using a little
of side information.
We understand that constrained spectral clustering (CSC)
algorithms [9], [10], [11], [12], [13], [14], [15], [16] are in general
better than other constrained clustering algorithms in terms of
accuracy, in part because of the high accuracies of unsupervised
spectral clustering [18], [19], [20], [21], [22]. However, the scal-
ability and efficiency of previous CSC algorithms are in general
poor. Specifically, the memory complexity of existing CSC algo-
rithms [9], [10], [11], [12], [13], [14], [15], [16] is Oðn2
Þ and the
time complexity is Oðn3
Þ, where n is the total number of instan-
ces. This hampers the applications of CSC algorithms towards
moderate and large datasets.
In this paper, we develop an efficient and scalable CSC algo-
rithm that can well handle moderate and large datasets. The
SCACS algorithm can be understood as a scalable version of the
well-designed but less efficient algorithm known as Flexible Con-
strained Spectral Clustering (FCSC) [15], [16]. To our best knowl-
edge, our algorithm is the first efficient and scalable version in this
area, which is derived by an integration of two recent studies, the
constrained normalized cuts [15], [16] and the graph construction
method based on sparse coding [23]. However, it is by no means
straightforward to integrate the two existing methods. In the rest
paper, we mainly answer three questions: how do we achieve an
effective integration? how do we derive the SCACS algorithm in
a principled way? and how well does the proposed algorithm
perform?
The structure of the rest paper is as follows: in Section 2, we
revisit the constrained normalized-cuts problem and the method of
graph construction based on sparse coding; in Section 3 we formu-
late the problem to be solved and present a closed-form mathemat-
ical analysis; in Section 4, we propose our algorithm based on the
results of Section 3; in Section 5, we evaluate the proposed algo-
rithm over benchmark datasets and discuss the results; In Section 6,
we draw the conclusion and mention the future work.
For notation, vectors and matrices are denoted by bold lower-
case and upper-case letters, respectively. Sets are denoted by italic
upper-case letters. Scalars are denoted by italic lower-case letters.
2 BACKGROUND
In this section, we revisit two pioneer studies, namely the con-
strained normalized cuts [15], [16] and the sparse coding based
graph construction method [23].
2.1 Constrained Normalized Cuts
Given a vector dataset X ¼ fxign
i¼1 where xi 2 Rd
, and a constraint
set fC¼; C6¼g where ðxi; xjÞ 2 C¼ if the patterns of xi and xj are
similar and ðxi; xjÞ 2 C6¼ otherwise, the aim is to partition X into k
clusters biased by the constraint set. Let W be the similarity matrix
over X where Wij represents the similarity between instances xi
and xj. Let D be the degree matrix over X which is a diagonal
matrix with elements Dii ¼
P
j Wij. Let L ¼ I À DÀ1=2
WDÀ1=2
be
the normalized graph Laplacian where I denotes the identity
matrix. Let Q denote the constraint matrix where Qij ¼ 1 expresses
ðxi; xjÞ 2 C¼, and Qij ¼ À1 expresses ðxi; xjÞ 2 C6¼, and Qij ¼ 0
expresses no available side information. Let Q ¼ DÀ1=2
QDÀ1=2
be
the normalized constraint matrix. The constrained normalized cuts
[15], [16] can be rewritten as:
arg minv2fþ 1ffiffin
p ;À 1ffiffin
p gn vT
Lv
s:t: vT
Qv ! a; vT
v ¼ 1; v?D1=2
1:
(1)
Here the parameter a controls what degree the input side informa-
tion is respected. After removed vT
Qv ! a, Eq. (1) degenerates to
the standard normalized cuts [18]. The problem is NP hard and the
feasible way is to allow v to take any real values [21] which is called
a relaxed method in the literature. The relaxed solution vector can
be given by generalized eigenvalue decomposition [15], [16]. How-
ever, it still requires Oðn2
Þ memory cost and Oðn3
Þ time cost.
2.2 Sparse Coding Based Graph Construction
Graph construction amounts to computing a similarity matrix.
There exist a variety of methods [18], [21], [23], [24], [25], [26], [27].
J. Li is with the College of Computer Science, Zhejiang University, Hangzhou
310058, China, and the Big Data Research Center of Enjoyor Co Ltd, HangZhou
310030, China. E-mail: jylbob@gmail.com.
Y. Xia is with the College of Computer Science, Zhejiang University, Hangzhou
310058, China. E-mail: xiayingjie@zju.edu.cn.
Z. Shan is with the College of Computer Science, Hangzhou Normal University,
Hangzhou 310012, China. E-mail: shanzhenyu@zju.edu.cn.
Y. Liu is with the Department of Automation, Shanghai Jiaotong University, Shang-
hai 200240, China. E-mail: company8110@gmail.com.
Manuscript received 1 Aug. 2013; revised 13 June 2014; accepted 18 Aug. 2014. Date of
publication 9 Sept. 2014; date of current version 23 Dec. 2014.
Recommended for acceptance by S. Chawla.
For information on obtaining reprints of this article, please send e-mail to: reprints@ieee.
org, and reference the Digital Object Identifier below.
Digital Object Identifier no. 10.1109/TKDE.2014.2356471
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 27, NO. 2, FEBRUARY 2015 589
1041-4347 ß 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
2. Among them, the method in [23] is based on sparse coding theory
[28] and is designed for handling large datasets.
Here we briefly introduce the sparse coding based graph con-
struction. Given a dataset X, of the form d-by-n matrix, X, sparse
coding aims to find a pair of matrices, U 2 RdÂp
and Z 2 RpÂn
,
such that UZ could best approximate X where U’s columns repre-
sent the desired base vectors and Z’s columns represent sparse coef-
ficient vectors—each vector has few non-zero components. The
cost function to be minimized is
fðU; ZÞ ¼ kX À UZk2
F : (2)
Unfortunately, it is costly to precisely solve for UÃ
and ZÃ
[29]. In
practice, an approximate and efficient way [23] is to randomly
choose p instances among the input dataset and act them as base
vectors, and then to estimate each column vector of Z according to
Zij ¼
Ksðxj; uiÞ
P
i2rNBðjÞ Ksðxj; uiÞ
; (3)
where s denotes the bandwidth of Gaussian kernel KsðÁ; ÁÞ and
i 2 rNBðjÞ means that the base vector ui is among the r nearest
base (rNB) vectors of instance xj. The sparseness of the matrix Z is
controlled by the parameter r.
After obtained Z, one can construct two forms of graph matrices:
G ¼ ZT
Z and S ¼ ZZT
. Choose ^Z ¼ DÀ1=2
Z where Dii ¼
P
j Zij. We
can obtain the normalized graph matrices ^G ¼ ^ZT ^Z 2 RnÂn
and
^S ¼ ^Z^ZT
2 RpÂp
. It is easy to check that the normalized graph Lapla-
cian over X is ðI À ^GÞ. The time complexity for computing ^G is
Oðpn2
Þ, but the one for computing ^S is just Oðnp2
Þ (p ( n).
3 PROBLEM FORMULATION
In this section, we formulate a scalable constrained normalized-
cuts problem and demonstrate how to solve it.
3.1 The Scalable Constrained Normalized Cuts
In the following, Problem 1 refers to a straightforward integration
of the constrained normalized cuts and the sparse coding based
graph construction, and Problem 2 refers to the formulated scalable
constrained normalized-cuts problem.
Problem 1. Let ^G be the similarity matrix over X. The relaxed con-
strained normalized-cuts problem can be formulated as
min
v2Rn
vT
Lv; s:t: vT
Qv ! a; vT
v ¼ 1; v?1; (4)
where L ¼ I À ^G is the normalized graph Laplacian.
Based on [15], the normal approach for finding the solution vec-
tor of Problem 1 can be reduced to the following generalized eigen-
value problem:
Lv ¼ ðQ À bIÞv; (5)
where b is a lower bound of a. The time cost for solving this prob-
lem is Oðn3
Þ, infeasible for handling large datasets.
To seek a scalable solution, we write v 2 Rn
as ^Z
T
u where
u 2 Rp
. Plugging v ¼ ^Z
T
u into Eq. (4), we reformulate Problem 1
as the following Problem 2.
Problem 2.
min
u2Rp
uT
Au; s:t: uT ^Qu ! a; uT ^Su ¼ 1; 1T ^Su ¼ 0; (6)
here,
A ¼ ^S À ^S^S; ^Q ¼ ^ZQ^Z
T
: (7)
This problem is mathematically equivalent to Problem 1, but
it results in two significant changes: (1) the n-by-n normalized
graph Laplacian L is compressed as the p-by-p matrix A; (2)
the n-by-n constraint matrix Q is naturally compressed as the
p-by-p matrix ^Q. Consequently, the solution of Problem 1 might
be efficiently recovered from the solution of Problem 2 consid-
ering p ( n.
3.2 Solving Problem 2
To solve Problem 2, we use Lagrange multiplier and obtain that
Lðu; ; mÞ ¼ uT
Au À ðuT ^Qu À aÞ À mðuT ^Su À 1Þ: (8)
Using the KKT Theorem [30], the feasible solutions must satisfy the
the following conditions:
Au À ^Qu À m^Su ¼ 0 (9)
uT ^Qu ! a (10)
uT ^Su ¼ 1; (11)
1T ^Su ¼ 0; (12)
! 0; (13)
ðuT ^Qu À aÞ ¼ 0: (14)
When ¼ 0, Eq. (9) degenerates to the standard eigen-system
^Su ¼ ð1 À mÞu, namely side information does not work. To use
side information, we limit 0 which reduces Eq. (10) and
Eq. (14) to
uT ^Qu ¼ a; (15)
Assuming that Àm= ¼ b, Eq. (9) can be reduced to
Au ¼ ð^Q À b^SÞu; (16)
which is a generalized eigenvalue problem that can be solved very
efficiently due to p ( n.
To derive a scalable constrained spectral clustering algorithm
(SCACS), it is essential to discuss how to set the parameters b and
a. Via a few algebraic manipulations, we find two facts which are
useful for algorithm design. First, the matrix A is positive semi-
definite, hence we have
uT
ð^Q À b^SÞu ¼ ða À bÞ ! 0; (17)
meaning that the parameter a is lower-bounded by b. Thus, users
just need to specify b, regardless of a. Second, for ensuring that
Eq. (16) has at least i meaningful solution vectors in terms of non-
negative normalized cuts, b gi is a sufficient condition where gi
denotes the ith largest eigenvalue of the generalized eigen-system
^Qx ¼ g^Sx.
4 ALGORITHM AND ANALYSIS
Based on the mathematical analysis above, in this section we derive
the algorithm and analyze the complexity.
4.1 The Proposed Algorithm
Problem 2 just indicates a binary constrained spectral clustering
problem where the solution vector vÃ
plays the role of group-
ing indicator. Without loss of generality, here we directly
derive an algorithm for k-class problems (k ! 2). We call the
proposed algorithm scalable constrained spectral clustering and
list 13 steps in Algorithm 1.
590 IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 27, NO. 2, FEBRUARY 2015
3. Algorithm 1. Scalable Constrained Spectral Clustering
1: Input: a dataset X 2 RdÂn
, the base-vector number p, the
n-by-n constraint matrix Q, b, cluster number k.
2: Choose p vector data among the input dataset at random,
and stack them in the columns of matrix U 2 RdÂp
.
3: Compute Z 2 RpÂn
using Eq. (3) and then compute
^Z ¼ DÀ1=2
Z where D denotes a diagonal matrix with
elements Dii ¼
P
j Zij.
4: Compute ^S ¼ ^Z^Z
T
and ^Q ¼ ^ZQ^Z
T
.
5: Find the largest eigenvalue gmax of the generalized eigen-
system ^Qx ¼ g^Sx.
6: If b ! gmax, return fvÃ
g ¼ ;; otherwise, find all the eigen-
vectors fuig by solving generalized eigen-system Eq. (16),
where ui denotes the ith eigenvector, 1 i p.
7: Find among fuig the eigenvectors fuigþ
associated with
positive eigenvalues.
8: Normalize each ui 2 fuigþ
by multiplying a factor
ffiffiffiffiffiffiffiffiffiffi
1
uT
i
^Sui
q
.
9: Remove the eigenvectors from fuigþ
that are not orthogonal
to the vector 1T ^S.
10: Find among fuigþ
the m eigenvectors that lead to the small-
est values of uT
i Aui where m ¼ minfk À 1; jfuigþ
jg, then
stack them in columns of matrix V.
11: Compute VðrÞ
¼ ^Z
T
VðI À VT
AVÞ.
12: Normalize VðrÞ
’s rows to have unit length, then feed it to the
k-means algorithm.
13: Output: the grouping indicator.
Several key steps are interpreted as follows: (1) Step 7 aims to
satisfy the condition 0; (2) Step 8 aims to scale each eigenvec-
tors for satisfying the condition of Eq. (11); (3) Step 9 aims to satisfy
the condition of Eq. (12); (4) In Step 11, we recover the solution vec-
tors by the linear transformation ^Z
T
u and we weight each solution
vector by one minus the associated value of the objective function.
It is worth mentioning that the input parameter b is tunable,
making the algorithm flexible to noisy side information or inappro-
priate mathematical expressions for side information. Usually, the
larger b is given, the more side information is respected.
4.2 Complexity Analysis
The algorithm complexity is analyzed as follows. The time cost for
computing the matrix ^S is in general Oðnp2
Þ, and that for comput-
ing the generalized eigenvalue decomposition in Step 5 or in Step 6
is Oðp3
Þ, and that for computing ^ZQ^ZT
is Oðkp2
þ kpnÞ. Hence the
general time complexity of our algorithm is
Oðkpn þ kp2
þ np2
þ p3
Þ: (18)
In practical applications, p can be chosen as a value far smaller than
n so that the running time grows almost linearly to n. The memory
cost for storing the matrices ^Z and ^S are OðnpÞ and Oðp2
Þ, respec-
tively. The general memory complexity is
Oðnp þ p2
Þ: (19)
5 EXPERIMENT
In this section, we carry out experiments for assessing the pro-
posed SCACS algorithm.
5.1 Experiment Setup
We implement our algorithm and other compared algorithms over
Linux machines, the machine for recording computational time
with 3.10 GHz CPU and 8 GB main memory.
The datasets that we used are downloaded from UCI machine
learning repository1
and the other two web sites.2;3
Basic informa-
tion is listed in Table 1.
Side information is generated based on the ground truth labels
of the datasets. The following expression illustrates the encoding
rules of Q:
Qij ¼
1 if xi and xj have consistent labels
1 if i ¼ j
À1 if xi and xj have different labels
0 no side information:
8
:
(20)
In our experiment, we randomly sample c labelled instances from a
given input dataset, and then obtain Q based on the rules of
Eq. (20).
The clustering accuracy is evaluated by the best matching rate
(ACC). Let h be the resulting label vector obtained from a cluster-
ing algorithm. Let g be the ground truth label vector. Then, the best
matching rate is defined as
ACC ¼
Pn
i¼1 dðgi; mapðhiÞÞ
n
; (21)
where dða; bÞ denotes the delta function that returns 1 if a ¼ b and
returns 0 otherwise, and map(hi) is the permutation mapping func-
tion that maps each cluster label hi to the equivalent label from the
data corpus.
5.2 Comparisons with Benchmarks and FCSC
In this part, we compare our SCACS algorithm with three spectral
algorithms (LSC-R, SL and FCSC). Among them, “LSC-R” [23] is
the unsupervised baseline that constructs similarity graphs based
on sparse coding. “SL” [10] is the CSC baseline which encodes side
information by directly modifying similarity matrices. “FCSC” [16]
is the state-of-the-art in terms of accuracy. FCSC is too slow to han-
dle large datasets, hence in this experiment we used four small
datasets:
1) The Image Segmentation dataset.
2) A subset sampled from the USPS dataset. We used the first
300 instances of each class, 3,000 instances in total.
3) A subset sampled from the Pen Digits dataset. We used the
first 300 instances of each class, 3000 instances in total.
4) A subset sampled from the Letter Recognition dataset. We
used the first five classes of instances, 3,864 instances in
total.
We set the Gaussian kernel width s to the average Euclidean dis-
tance between all instances and base vectors. We use p ¼ 500, r ¼ 3
and b ¼ b0gkÀ1 where b0 ¼ 0:5 þ 0:4 Â c
n and gkÀ1 denotes the (k-1)
th largest eigenvalue of the generalized eigen-system ^Qx ¼ g^Sx.
For fair comparison, the randomly selected labelled instances are
TABLE 1
Dataset Description
Dataset # Instances # Attributes # Classes
Image Segmentation 2,310 19 7
USPS 9,298 256 10
Pen Digits 10,992 16 10
Letter Recognition 20,000 16 26
MNIST 70,000 784 10
CoverType 581,012 54 7
1. http://archive.ics.uci.edu/ml
2. http://www.zjucadcg.cn/dengcai
3. http://yann.lecun.com/exdb/mnist
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 27, NO. 2, FEBRUARY 2015 591
4. consistent for SL, SCACS and FCSC. In order to show how the
results vary with c, we use 10 different values ranging from 100 to
1,000 by a increasing step size 100; for each value, we carry out 20 tri-
als in terms of 20 different constraint sets. For SL, we use the k-near-
est neighbor method to compute similarity matrices (k ¼ 30), and
we use radial basis function (RBF) to compute the similarities
between pairwise instances. For FCSC, we directly use RBF kernel
function to compute similarity matrices as used by the authors.
We demonstrate the algorithm accuracies and the standard devi-
ations in Figs. 1a, 1b, 1c, and 1d. One can see that our algorithm out-
performs LSC-R over all the datasets with large margins, indicating
that our algorithm in encoding side information is appropriate. In
addition, FCSC significantly outperforms SL, indicating that the
constrained normalized-cuts framework performs much better in
encoding side information. Over the four small datasets, our algo-
rithm outperforms SL on average and is close to FCSC.
5.3 Comparisons with Efficient Algorithms
In this part, we compare our algorithm with three efficient algo-
rithms (LSC-R, CKM, and ML). Among them, “CKM” refers to the
k-means based constrained clustering algorithm [2]. “ML” refers to
the side information based metric learning method [4] and we use
their efficient version that learns a diagonal matrix. The datasets
that we used are listed in Table 1.
For fair comparison, we use the same base-vector selections for
LSC-R and SCACS, and we use the same random selections of con-
strained instances for CKM, ML and SCACS. We carry out 20 trials
in terms of 20 different constrained instance sets, and report the
average clustering accuracies and the standard deviations. The
parameter that we used are p ¼ 500, b0 ¼ 0:1, r ¼ 3, respectively.
The results are illustrated in Figs. 1e, 1f, 1g, 1h, 1i, and 1j. Over
five datasets (Image Segmentation, USPS, Pen Digits, LetterRec
and MNIST), our algorithm significantly outperforms other three
algorithms; and over the remaining CoverType dataset, our algo-
rithm performs better than LSC-R and CKM. Note that ML per-
forms well on low-dimensional datasets such as Letter Recognition
and CoverType, with a decreased performance on high-dimen-
sional datasets such as USPS and MNIST. However, our algorithm
performs much better on high-dimensional datasets.
The running time is recorded in Table 2. The results show that
SCACS is much faster than FCSC. For the Pen Digits dataset, FCSC
demands more than 11 hours, however, our algorithm demands
only 3.22 seconds. For the two largest datasets, MNIST and Cov-
erType, FCSC could not return results within a week, however, our
algorithm only demands 12.02 and 64.47 seconds, respectively.
5.4 Influence of Parameters
Below we show how the parameters (r, p, b0) influence SCACS. We
use two datasets, Pen Digits and MNIST, with a lower dimension-
ality and a higher dimensionality, respectively. For computing the
influence of r, we fix c ¼ 500 and vary r from 3 to 30 by a step size
1. For computing the influence of p, we fix c ¼ 500 and vary p rang-
ing from 100 to 1,000 by a step size 100. For computing the influ-
ence of b0, we vary b0 from 0.1 to 0.9 by a step size 0.1.
Fig. 2a shows the influence of r. The accuracies of SCACS tend
to decrease as r increases, indicating that sparseness plays a key
role in our algorithm. This suggests that a relative small r is prefer-
able. The accuracy decrease pace over the MNIST dataset is faster
than that over the Pen Digits dataset. This might support that
sparseness is more important for grouping high-dimensional data-
sets than for grouping low-dimensional datasets. Fig. 2b shows
that the accuracies of SCACS vary mildly with p; in particular in
the interval of, e.g., p ranging from 500 to 1,000, our algorithm is
significantly robust. This states that it is easy to set a usable value
for p. Figs. 2c and 2d show how the clustering accuracies vary with
b0. One may notice that the clustering accuracies drop seriously if
constrained instances are very few (e.g., c ¼ 100) but b0 uses a
larger value (e.g., b0 ¼ 0.8). Conversely, if constrained instances are
abundant, a larger b0 might result in a significant improvement in
terms of accuracy, e.g., for the Pen Digits datasets, the clustering
accuracies can be significantly improved as b0 increases when
c ¼ 5;000.
6 CONCLUSION AND FUTURE WORK
We have developed a new k-way scalable constrained spectral
clustering algorithm based on a closed-form integration of the con-
strained normalized cuts and the sparse coding based graph con-
struction. Experimental results show that (1) with less side
information, our algorithm can obtain significant improvements in
Fig. 1. Accuracy comparisons among different algorithms. Here c denotes the number of labelled data.
TABLE 2
Running Time (seconds), c ¼ 100, p ¼ 500
Dataset LSC-R CKM ML SCACS FCSC
ImageSeg 0.41 0.01 3.46 2.35 153
USPS 0.96 0.82 21.05 3.29 14,461
PenDigits 0.82 0.31 5.42 3.22 39,948
LetterRec 4.12 4.10 15.17 6.59 212,600
MNIST 9.32 56.70 492.9 12.02 -
CoverType 49.12 33.31 190.58 64.47 -
592 IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 27, NO. 2, FEBRUARY 2015
5. accuracy compared to the unsupervised baseline; (2) with less
computational time, our algorithm can obtain high clustering accu-
racies close to those of the state-of-the-art; (3) It is easy to select the
input parameters; (4) our algorithm performs well in grouping
high-dimensional image data. In the future, we are considering an
active selection of pairwise instances for labelling; we will also
apply our algorithm to group urban transportation big data, which
might significantly boost sensor placement optimization.
ACKNOWLEDGMENTS
This paper was supported in part by the following funds: National
High Technology Research and Development Program of China
(2011AA010101), National Natural Science Foundation of China
(61002009 and 61304188), Key Science and Technology Program of
Zhejiang Province of China (2012C01035-1), and Zhejiang Provin-
cial Natural Science Foundation of China (LZ13F020004 and
LR14F020003), Postdoctoral fund of China. The authors would like
to thank Prof. Xiaofei He for beneficial suggestions. Yingjie Xia is
the corresponding author.
REFERENCES
[1] K. Wagstaff and C. Cardie, “Clustering with instance-level constraints,” in
Proc. 17th Int. Conf. Mach. Learn., 2000, pp. 1103–1110.
[2] K. Wagstaff, C. Cardie, S. Rogers, and S. Schroedl, “Constrained k-means
clustering with background knowledge,” in Proc. 18th Int. Conf. Mach.
Learn., 2001, pp. 577–584.
[3] S. Basu, A. Banerjee, and R. Mooney, “Semi-supervised clustering by
seeding,” in Proc. 19th Int. Conf. Mach. Learn., 2002, pp. 27–34.
[4] E. P. Xing, A. Y. Ng, M. I. Jordan, and S. J. Russell, “Distance metric learn-
ing with application to clustering with side-information,” in Proc. Adv. Neu-
ral Inf. Process. Syst., 2003, pp. 505–512.
[5] S. Basu, B. I. Lenko, and R. J. Mooney, “A probabilistic framework for semi-
supervised clustering,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data
Mining, 2004, pp. 59–68.
[6] N. Shental, A. Bar-hillel, T. Hertz, and D. Weinshall, “Computing gaussian
mixture models with em using equivalence constraints,” in Proc. Adv. Neu-
ral Inf. Process. Syst. 16, 2003, pp. 505–512.
[7] B. Kulis, S. Basu, I. Dhillon, and R. Mooney, “Semi-supervised graph clus-
tering: A kernel approach,” in Proc. 22nd Int. Conf. Mach. Learn., 2005,
pp. 457–464.
[8] Y.-M. Cheung and H. Zeng, “Semi-supervised maximum margin clustering
with pairwise constraints,” IEEE Trans. Knowl. Data Eng., vol. 24, no. 5,
pp. 926–939, May 2012.
[9] S. X. Yu and J. B. Shi, “Grouping with bias,” in Proc. Adv. Neural Inf. Process.
Syst., 2001, pp. 1327–1334.
[10] S. D. Kamvar, D. Klein, and C. D. Manning, “Spectral learning,” in Proc. Int.
Joint Conf. Artif. Intell., 2003, pp. 561–566.
[11] X. Ji and W. Xu, “Document clustering with prior knowledge,” in Proc.
29th Annu. Int. Conf. Res. Develop. Inf. Retrieval, 2006, pp. 405–412.
[12] Z. Lu and M. A. Carreira-Perpinan, “Constrained spectral clustering
through affinity propagation,” in Proc. IEEE Conf. Comput. Vis. Pattern Rec-
ognit., 2008, pp. 1–8.
[13] Z. Li, J. Liu, and X. Tang, “Constrained clustering via spectral regu-
larization,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit., 2009, pp. 421–
428.
[14] T. Coleman, J. Saunderson, and A. Wirth, “Spectral clustering with incon-
sistent advice,” in Proc. 25th Int. Conf. Mach. Learn., 2008, pp. 152–159.
[15] X. Wang and I. Davidson, “Flexible constrained spectral clustering,” in
Proc. 16th ACM SIGKDD Int. Conf. Knowl. Discov. Data Mining, 2010,
pp. 563–572.
[16] X. Wang, B. Qian, and I. Davidson, “On constrained spectral clustering and
its applications,” Data Mining Knowl. Discov., vol. 28, no. 1, pp. 1–30, 2012.
[17] M. Schultz and T. Joachims, “Learning a distance metric from relative
comparisons,” in Proc. Adv. Neural Inf. Process. Syst., 2003, pp. 41–48.
[18] J. B. Shi and J. Malik, “Normalized cuts and image segmentation,” IEEE
Trans. Pattern Anal. Mach. Intell., vol. 22, no. 8, pp. 888–905, Aug. 2000.
[19] A. Y. Ng, M. I. Jordan, and Y. Weiss, “On spectral clustering: Analysis and
an algorithm,” in Proc. Adv. Neural Inf. Process. Syst. 14, 2001, pp. 849–856.
[20] C. Fowlkes, S. Belongie, F. R. K. Chung, and J. Malik, “Spectral grouping
using the Nystr€om method,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 26,
no. 2, pp. 214–225, Feb. 2004.
[21] U. V. Luxburg, “A tutorial on spectral clustering,” Statist. Comput., vol. 17,
no. 4, pp. 395–416, 2007.
[22] Y. Yang, Y. Yang, H. Shen, Y. Zhang, X. Du, and X. Zhou,
“Discriminative nonnegative spectral clustering with out-of-sample
extension,” IEEE Trans. Knowl. Data Eng., vol. 25, no. 8, pp. 1760–1771,
Aug. 2013.
[23] X. Chen and D. Cai, “Large scale spectral clustering with landmark-based
representation,” in Proc. 25th Conf. Artif. Intell., 2011, p. 1.
[24] S. I. Daitch, J. A. Kelner, and D. A. Spielman, “Fitting a graph to vector
data,” in Proc. 26th Annu. Int. Conf. Mach. Learn., 2009, p. 26.
[25] T. Jebara J. Wang, and S. Chang, “Graph construction and b-matching for
semi-supervised learning,” in Proc. Int. Conf. Mach. Learn., 2009, p. 56.
[26] B. Cheng, J. Yang S. Yan Y. Fu, and T. S. Huang, “Learning with l1-graph
for image analysis,” IEEE Trans. Image Process., vol. 19, no. 4, pp. 858–866,
Apr. 2010.
[27] W. Liu, J. He, and S. Chang, “Large graph construction for scalable semi-
supervised learning,” in Proc. Int. Conf. Mach. Learn., 2010, pp. 679–686.
[28] B. A. Olshausen, “Emergence of simple-cell receptive field properties by
learning a sparse code for natural images,” Nature, vol. 381, pp. 607–609,
1996.
[29] H. Lee A. Battle R. Raina, and A. Y. Ng, “Efficient sparse coding algo-
rithms,” in Proc. Adv. Neural Inf. Process. Syst. 19, 2006, pp. 801–808.
[30] H. W. Kuhn and A. W. Tucker, “Nonlinear programming,” in Proc. 2nd Ber-
keley Symp., 1951, pp. 481–492.
For more information on this or any other computing topic,
please visit our Digital Library at www.computer.org/publications/dlib.
Fig. 2. The influence of parameters (r,p,b0) to the proposed algorithm.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, VOL. 27, NO. 2, FEBRUARY 2015 593