0
Mobile Ontology Cloud- Semantic Post-IT-<br />IT Life and Ontology<br />Key-Sun Choi (kschoi@kaist.edu)<br />http://kschoi...
1st day: what we will learn<br />)<br /><ul><li>What is Semantic Post-it? (15 min)
Demo and Downloadable (5 min)
Enabling Technologies (15 min)
APIs for Technologies (5 min)
ontocore.org (what you can do),
Protégé API
Remaining in your home
References to read and to use </li></li></ul><li>What is Semantic Post-it?: Contents<br /><ul><li>As Mobile App
Personal Ontology Editors
Benefits when interpreting the input messages</li></li></ul><li>Introduction<br />What is the Semantic Post-It?<br /><ul><...
Personal randomized message
Organizing by interpreting messages
Table information extraction from text
Relevant table information grouping
Personal information space
Usage of ontology that user can edit
Collective intelligence
Usage of pivot ontology based on Wikipedia (web-based encyclopedia that anyone can edit) </li></li></ul><li>Windows Mobile...
Motivation<br />Motivating Scenario<br />Reading an article on “Omnia 2”<br />Another similar Smartphone? <br />More detai...
Motivation<br />Motivating Scenario<br />CPU clock<br />Reading an article on “Omnia 2”<br />Another similar Smartphone? <...
Motivation<br />Motivating Scenario<br />nationality<br />Reading an article on “Immanuel Kant”<br />Where is he from?<br ...
Motivation<br />What is the solution?<br /><ul><li>We need a system that retrieves relevant information
Data set that specifies attributes for each concepts is needed
Smartphone : manufacturer, OS, memory, …
Philosophers : nationality, follower, teacher, …
However, no one guy can describe every concepts
We can obtain the data set from collective intelligence</li></ul>author<br />politician<br />scientist<br />engineer<br />...
Established<br />February 16, 1971<br />Type<br />Government-run<br />President<br />Nam-Pyo Suh<br />…<br />…<br />Motiva...
Background Technologies<br />New paradigm<br /><ul><li>A few years have passed since a new paradigm was introduced.
Semantic Web
A machine-readable web
Ontology
A formal specification of knowledge</li></li></ul><li>Background Technologies<br />Semantic Web<br /><ul><li>An evolving d...
The meaning (semantics) of information and services on the web is defined
For the web to "understand" and satisfy the requests of people and machines to use the web content</li></ul>Our focus<br /...
Background Technologies<br />RDF<br /><ul><li>Resource Description Framework</li></ul>A Wikipedia article about Tony Benn<...
Background Technologies<br />Ontology<br />A formal specification of knowledge to be interpreted by computers<br />Company...
Illustrative Example<br />Content Space -> Message Space<br />Semantic Post-It<br />(Message List)<br />Typical Web Browse...
Illustrative Example<br />Message Space -> Triple Message Space (1/2)<br />Semantic Post-It<br />(Message List)<br />Seman...
Estabilshed<br />1971<br />Province<br />Daejeon<br />Country<br />South Korea<br />…<br />…<br />Illustrative Example<br ...
Illustrative Example<br />Triple Message Space ->Linked Triple Message Space<br />Semantic Post-It<br />(Graph View)<br />...
Illustrative Example<br />Linked Triple Message Space<br />Semantic Post-It<br />(Using Ontology 1)<br />Semantic Post-It<...
Illustrative Example<br />Personal Ontology Editor<br /><ul><li>Rename the property name
If you wish to see another label in the link
Ex) isManufacturedBy -> manufacturer
Modify constraints
If you wish to see the country name rather than the city name
Ex)
Remove : University-province-Settlement
Add : University-locatedAt-Country
Use the modified ontology in your Semantic Post-It</li></ul>How to embed this complex UI into Smartphone?<br />http://prot...
System architecture (1/2)<br />Message Interpretation Services<br />HTTP request<br />Semantic Post-IT Server<br />(HTTP s...
System architecture (2/2)<br />Semantic Post-IT client<br /><ul><li>Local Message DB controller
Message input interface
Message list viewer
HTTP service controller
Semantic Post-IT server
External message service
Message relation graph viewer
Personal ontology editor</li></ul>Semantic Post-IT client<br />Personal Ontology<br />Local Message DB<br />
Demo and Downloadable<br /> http://swrc.kaist.ac.kr/SemanticToolkits/<br />
What is Semantic Post-It?<br />Memo Admin Service <br />Evernote, quickies, etc.<br />Semantic Service Mash-Up<br />
Semantic Service Mash-up<br />Definition of 3 types of applications<br />Type 1 Application: Information zooming on specif...
Type 1 Application: Example<br />DEMO: Semantic Post-It<br />
Type 2 Application: Demo<br />DEMO: Semantic Post-It<br />
Type 3 Application: Demo<br />DEMO: Semantic Post-It<br />
Structure of Semantic Post-It<br />Post-It Server<br />Post-It Client<br />Service<br />Repository<br />Communication betw...
Support for Semantic Post-It:OntoCloud<br />Ontology derived from Wikipedia infoboxes<br />Official Website:http://swrc.ka...
Support for Type 2 Application:Semantic Annotation<br />One of possible type 2 application: Table-form summary generator<b...
Semantic Annotation Toolkit: COAT<br />DEMO: COAT<br />
From annotated data to Application: Machine Learning Feature<br />Support Vector Machine(SVM)<br />
Ontology Feature<br /> Modern GSM-based BlackBerry handhelds incorporate an ARM 7 or 9 processor, while older BlackBerry 9...
Data Authority Policy<br />Annotators can check his/her documents ONLY!<br />To prevent cheating<br />Simple annotation da...
Support for Type 3 Application:300M Wikipedia articles into Database<br />Provide baseline for shared memo<br />For type 3...
Screenshots<br />3) Table information extraction<br />1) User inputs message<br />2) Ontology recommendation<br />4) Relev...
Enabling Technologies<br /><ul><li>CAT2ISA
Table Generator  </li></li></ul><li>Technologies<br />Ontology expression<br />OWL (Web Ontology Language)<br /><owl:Class...
Technologies<br />Ontology inference<br />Text<br />Text<br />Samsung releases Omnia<br />Text<br />Apple releases <br />i...
Upcoming SlideShare
Loading in...5
×

2/24(Wed) - PowerPoint Presentation

1,429

Published on

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
1,429
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
11
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Transcript of "2/24(Wed) - PowerPoint Presentation"

  1. 1. Mobile Ontology Cloud- Semantic Post-IT-<br />IT Life and Ontology<br />Key-Sun Choi (kschoi@kaist.edu)<br />http://kschoi.kaist.ac.kr/ <br />CILab & Semantic Web Research<br />
  2. 2. 1st day: what we will learn<br />)<br /><ul><li>What is Semantic Post-it? (15 min)
  3. 3. Demo and Downloadable (5 min)
  4. 4. Enabling Technologies (15 min)
  5. 5. APIs for Technologies (5 min)
  6. 6. ontocore.org (what you can do),
  7. 7. Protégé API
  8. 8. Remaining in your home
  9. 9. References to read and to use </li></li></ul><li>What is Semantic Post-it?: Contents<br /><ul><li>As Mobile App
  10. 10. Personal Ontology Editors
  11. 11. Benefits when interpreting the input messages</li></li></ul><li>Introduction<br />What is the Semantic Post-It?<br /><ul><li>A system that maps personal randomized message into well-organized personal information space based on collective intelligence.
  12. 12. Personal randomized message
  13. 13. Organizing by interpreting messages
  14. 14. Table information extraction from text
  15. 15. Relevant table information grouping
  16. 16. Personal information space
  17. 17. Usage of ontology that user can edit
  18. 18. Collective intelligence
  19. 19. Usage of pivot ontology based on Wikipedia (web-based encyclopedia that anyone can edit) </li></li></ul><li>Windows Mobile<br />isDevelopedBy<br />Microsoft<br />Windows Mobile<br />isDevelopedBy<br />Microsoft<br />Omnia 2<br />ISA<br />smartphone<br />Omnia 2<br />ISA<br />smartphone<br />Flash memory <br />ISA<br />computer storage <br />Flash memory <br />ISA<br />computer storage <br />Omnia 2<br />hasOS<br />Windows Mobile<br />Omnia 2<br />hasOS<br />Windows Mobile<br />Omnia 2<br />hasMemory<br />Flash memory<br />Omnia 2<br />hasMemory<br />Flash memory<br />Introduction<br />A working flow of Semantic Post-It<br />Omnia 2 is a multimedia smartphone announced at Samsung. Omnia 2 runs Windows Mobile and comes with flash memory..<br />Contents Space<br />Flash memory is a non-volatile computer storage that <br />can be electrically erased and reprogrammed.<br />Windows Mobile is a compact mobile operating system developed by Microsoft<br />Omnia 2 is a multimedia smartphone announced at Samsung. Omnia 2 runs Windows Mobile and comes with flash memory..<br />Message Space<br />Triple Message Space<br />(Table information)<br />hasMemory<br />hasOS<br />Linked Triple Message Space<br />
  20. 20. Motivation<br />Motivating Scenario<br />Reading an article on “Omnia 2”<br />Another similar Smartphone? <br />More details on OS<br />What is the recent trend of it?<br />Company in competition<br />What should we do?<br />
  21. 21. Motivation<br />Motivating Scenario<br />CPU clock<br />Reading an article on “Omnia 2”<br />Another similar Smartphone? <br />OS, platform<br />More details on OS<br />What is the recent trend of it?<br />Manufacturer, design<br />Company in competition<br />Products of the company<br />We have to think of what type of information are involved<br />
  22. 22. Motivation<br />Motivating Scenario<br />nationality<br />Reading an article on “Immanuel Kant”<br />Where is he from?<br />?<br />?<br />?<br />If new to philosophers, <br />we are likely to have no idea about relevant information<br />
  23. 23. Motivation<br />What is the solution?<br /><ul><li>We need a system that retrieves relevant information
  24. 24. Data set that specifies attributes for each concepts is needed
  25. 25. Smartphone : manufacturer, OS, memory, …
  26. 26. Philosophers : nationality, follower, teacher, …
  27. 27. However, no one guy can describe every concepts
  28. 28. We can obtain the data set from collective intelligence</li></ul>author<br />politician<br />scientist<br />engineer<br />Philosophers<br />Artist<br />
  29. 29. Established<br />February 16, 1971<br />Type<br />Government-run<br />President<br />Nam-Pyo Suh<br />…<br />…<br />Motivation<br />Wikipedia<br />Wikipedia documents (2010/01/29)<br />3,175,836 (ENG) - 11,527,437 users<br />125,801 (KOR) - 100,498 users<br />④<br />①<br />②<br />① Inter-page link<br />② Inter-Language link<br />③ Category<br />④ Infobox: table information<br />③<br />
  30. 30. Background Technologies<br />New paradigm<br /><ul><li>A few years have passed since a new paradigm was introduced.
  31. 31. Semantic Web
  32. 32. A machine-readable web
  33. 33. Ontology
  34. 34. A formal specification of knowledge</li></li></ul><li>Background Technologies<br />Semantic Web<br /><ul><li>An evolving development of the World Wide Web
  35. 35. The meaning (semantics) of information and services on the web is defined
  36. 36. For the web to "understand" and satisfy the requests of people and machines to use the web content</li></ul>Our focus<br />Adapted from Wikipedia<br />(http://en.wikipedia.org/wiki/Semantic_Web)<br />
  37. 37. Background Technologies<br />RDF<br /><ul><li>Resource Description Framework</li></ul>A Wikipedia article about Tony Benn<br /><http://en.wikipedia.org/wiki/Tony_Benn> <http://purl.org/dc/elements/1.1/title> "Tony Benn" .<br /><http://en.wikipedia.org/wiki/Tony_Benn> <http://purl.org/dc/elements/1.1/publisher> "Wikipedia" .<br /><rdf:RDF<br />xmlns:rdf=http://www.w3.org/1999/02/22-rdf-syntax-ns#<br />xmlns:dc=http://purl.org/dc/elements/1.1/><br /><rdf:Descriptionrdf:about=http://en.wikipedia.org/wiki/Tony_Benn><br /><dc:title>Tony Benn</dc:title><br /><dc:publisher>Wikipedia</dc:publisher><br /></rdf:Description><br /></rdf:RDF><br />An expression of “triple”<br />Adapted from Wikipedia<br />(http://en.wikipedia.org/wiki/Resource_Description_Framework)<br />
  38. 38. Background Technologies<br />Ontology<br />A formal specification of knowledge to be interpreted by computers<br />Company<br />isManufacturedBy<br />supportSoftware<br />OS<br />releaseDate<br />rdfs:subPropertyOf<br />runsOn<br />cameraPixelOf<br />Mobile Phone<br />hasMemorySize<br />supportOnlineSoftware<br />hasWebsite<br />Mobile PhoneSoftware<br />rdfs:subClassOf<br />rdfs:subClassOf<br />rdfs:subClassOf<br />PDA<br />Cellular<br />Phone<br />SmartPhone<br />Cellular<br />Phone<br />Smart Phone<br />PDA<br />Schema<br />Instance<br />releaseDate<br />2008<br />Samsung<br />i900 Omnia<br />cameraPixelOf<br />5 megapixels<br />supportOnlineSoftware<br />hasWebsite<br />Skype<br />www.skype.com<br />hasMemorySize<br />128 MB<br />runsOn<br />Windows<br />Mobile 6.1<br />isManufacturedBy<br />Samsung<br />
  39. 39. Illustrative Example<br />Content Space -> Message Space<br />Semantic Post-It<br />(Message List)<br />Typical Web Browser<br />External Contents<br />KAIST is located in Daejeon, South Korea. KAIST was established by Korean government in 1971 <br />Scrap<br />Related Problems : Mash-Up<br />How to extract text from heterogeneous contents (in a context, not a scientific issue)<br />
  40. 40. Illustrative Example<br />Message Space -> Triple Message Space (1/2)<br />Semantic Post-It<br />(Message List)<br />Semantic Post-It<br />(Detail View)<br />Semantic Post-It<br />(Detail View)<br />KAIST is located in Daejeon, South Korea. KAIST was established by Korean government in 1971 <br />KAIST is located in Daejeon, South Korea. KAIST was established by Korean government in 1971.… The current KAIST President Nam Pyo Suhtaught for…<br />KAIST is located in Daejeon, South Korea. KAIST was established by Korean government in 1971.… The current KAIST President Nam Pyo Suhtaught for…<br />Person<br />Smatphone’s UI is limited. Information should be shown by one-click.<br />Related Problems : ISA relation recognition<br />
  41. 41. Estabilshed<br />1971<br />Province<br />Daejeon<br />Country<br />South Korea<br />…<br />…<br />Illustrative Example<br />Message Space -> Triple Message Space(2/2)<br />Semantic Post-It<br />(Table View)<br />Semantic Post-It<br />(Message List)<br />Semantic Post-It<br />(Message View)<br />KAIST is located in Daejeon, South Korea. KAIST was established by Korean government in 1971 <br />KAIST is located in Daejeon, South Korea. KAIST was established by Korean government in 1971.… The current KAIST President Nam Pyo Suhtaught for…<br />KAIST<br />Summarization<br />Display size is too small to do full browsing.<br />Related Problems : Triple extraction from text<br />
  42. 42. Illustrative Example<br />Triple Message Space ->Linked Triple Message Space<br />Semantic Post-It<br />(Graph View)<br />Semantic Post-It<br />(Message List)<br />Semantic Post-It<br />(Message View)<br />Suh was born in Korea on April 22, 1936, and immigrated to the U.S. in 1954….<br />KAIST is located in Daejeon, South Korea. KAIST was established by Korean government in 1971 <br />KAIST is located in Daejeon, South Korea. KAIST was established by Korean government in 1971.… The current KAIST President Nam Pyo Suhtaught for…<br />president<br />KAIST is located in Daedeok…<br />province<br />Daejeon is a center of transportation in South Korea, where two major,<br />Relevant messages<br />Display size is too small to show text<br />Related Problems : Relevant keyword search by traversing Ontology<br />
  43. 43. Illustrative Example<br />Linked Triple Message Space<br />Semantic Post-It<br />(Using Ontology 1)<br />Semantic Post-It<br />(Using Ontology 2)<br />Ontology 1<br />Ontology 2<br />Suh was born in Korea on April 22, 1936, and immigrated to the U.S. in 1954….<br />Suh was born in Korea on April 22, 1936, and immigrated to the U.S. in 1954….<br />University<br />University<br />president<br />president<br />president<br />president<br />KAIST is located in Daedeok…<br />KAIST is located in Daedeok…<br />Person<br />Person<br />province<br />province<br />province<br />locatedAt<br />Settlement<br />Country<br />Daejeon is a center of transportation in South Korea, where two major,<br />South Korea is a presidential republic consisting of 16 administrative…<br />Related Problems : Personal ontology editing, logical consistency checking<br />
  44. 44. Illustrative Example<br />Personal Ontology Editor<br /><ul><li>Rename the property name
  45. 45. If you wish to see another label in the link
  46. 46. Ex) isManufacturedBy -> manufacturer
  47. 47. Modify constraints
  48. 48. If you wish to see the country name rather than the city name
  49. 49. Ex)
  50. 50. Remove : University-province-Settlement
  51. 51. Add : University-locatedAt-Country
  52. 52. Use the modified ontology in your Semantic Post-It</li></ul>How to embed this complex UI into Smartphone?<br />http://protege.stanford.edu/<br />
  53. 53. System architecture (1/2)<br />Message Interpretation Services<br />HTTP request<br />Semantic Post-IT Server<br />(HTTP server)<br />Semantic Post-IT client<br />TABLEGEN<br />CAT2ISA<br />HTTP response<br />Ontology Access<br />DBpedia Access<br />Personal Ontology<br />Local Message DB<br />External Message Service<br />System Message DB<br />Twitter, Blog, Email, Calendar, …<br />
  54. 54. System architecture (2/2)<br />Semantic Post-IT client<br /><ul><li>Local Message DB controller
  55. 55. Message input interface
  56. 56. Message list viewer
  57. 57. HTTP service controller
  58. 58. Semantic Post-IT server
  59. 59. External message service
  60. 60. Message relation graph viewer
  61. 61. Personal ontology editor</li></ul>Semantic Post-IT client<br />Personal Ontology<br />Local Message DB<br />
  62. 62. Demo and Downloadable<br /> http://swrc.kaist.ac.kr/SemanticToolkits/<br />
  63. 63. What is Semantic Post-It?<br />Memo Admin Service <br />Evernote, quickies, etc.<br />Semantic Service Mash-Up<br />
  64. 64. Semantic Service Mash-up<br />Definition of 3 types of applications<br />Type 1 Application: Information zooming on specific ‘word’ of a memo<br />Type 2 Application: Memo Contents Analysis<br />Type 3 Application: Information zooming on whole context of a memo<br />
  65. 65. Type 1 Application: Example<br />DEMO: Semantic Post-It<br />
  66. 66. Type 2 Application: Demo<br />DEMO: Semantic Post-It<br />
  67. 67. Type 3 Application: Demo<br />DEMO: Semantic Post-It<br />
  68. 68. Structure of Semantic Post-It<br />Post-It Server<br />Post-It Client<br />Service<br />Repository<br />Communication between<br />Server and Client<br />Provide application List<br />Application Install <br />Request for new application<br />Execute<br />application<br />Request<br />Ontology<br />Ontology Request Module<br />Enterprise Part:<br />Add-on of <br />Semantic Applications<br />Shared Memo Request Module<br />Return shared memos which the client have requested<br />Can download shared memo to local database<br />Request<br />Shared Memo<br />Add new memo<br />Delete memo<br />Find<br />Related <br />Memo<br />Change memo<br />Synchronization Module<br /><ul><li>Synchronization between Server & Client</li></ul>Tag memo<br />Synchroni-zation<br />Attach ontology to memo<br />Shared <br />Memo<br />Ontology<br />Repository<br />Local File System<br />Personal<br />Memos<br />Wikipedia Documents<br />PURE PART<br />
  69. 69. Support for Semantic Post-It:OntoCloud<br />Ontology derived from Wikipedia infoboxes<br />Official Website:http://swrc.kaist.ac.kr/ontocloud/<br />
  70. 70. Support for Type 2 Application:Semantic Annotation<br />One of possible type 2 application: Table-form summary generator<br />Semantic Annotation: Mark on the documents – ‘which part’ could be transformed into table?<br />
  71. 71. Semantic Annotation Toolkit: COAT<br />DEMO: COAT<br />
  72. 72. From annotated data to Application: Machine Learning Feature<br />Support Vector Machine(SVM)<br />
  73. 73. Ontology Feature<br /> Modern GSM-based BlackBerry handhelds incorporate an ARM 7 or 9 processor, while older BlackBerry 950 and 957 handheldsused Intel 80386 processors.<br />IT Ontology Package<br />Gathering semantic Info<br />Using Ontology<br />CPU<br />Intel 80386<br /> Modern GSM-based BlackBerry handhelds incorporate an ARM 7 or 9 processor, while older BlackBerry 950 and 957 handheldsused Intel 80386 processors.<br />useCPU<br />
  74. 74. Data Authority Policy<br />Annotators can check his/her documents ONLY!<br />To prevent cheating<br />Simple annotation data viewer is available<br />For administrators<br />DEMO: COAT Viewer<br />
  75. 75. Support for Type 3 Application:300M Wikipedia articles into Database<br />Provide baseline for shared memo<br />For type 3 application<br />Build shared memo database with 300M wikipedia articles as its part<br />
  76. 76. Screenshots<br />3) Table information extraction<br />1) User inputs message<br />2) Ontology recommendation<br />4) Relevant message grouping<br />
  77. 77. Enabling Technologies<br /><ul><li>CAT2ISA
  78. 78. Table Generator  </li></li></ul><li>Technologies<br />Ontology expression<br />OWL (Web Ontology Language)<br /><owl:Class rdf:ID=“Mobile Phone"/> <br /><owl:Class rdf:ID=“PDA"><br /><rdfs:subClassOf rdf:resource=“# Mobile Phone"/><br /></owl:Class><br /><owl:Class rdf:ID=“SmartPhone"><br /><rdfs:subClassOf rdf:resource="# Mobile Phone"/><br /></owl:Class><br /><owl:Class rdf:ID=“Cellular Phone"><br /><rdfs:subClassOf rdf:resource="# Mobile Phone"/><br /></owl:Class><br /><owl:Class rdf:ID=“Mobile Phone Software"/><br /><owl:ObjectProperty rdf:ID=“hasSoftware"><br /><rdfs:domain rdf:resource="#Mobile Phone”/><br /><rdfs:range rdf:resource=“# Mobile Phone Software"/><br /></owl:ObjectProperty><br /><owl:ObjectProperty rdf:ID=“hasOnlineSoftware"><br /><rdfs:subPropertyOf rdf:resource=“#hasSoftware"/><br /></owl:ObjectProperty><br />
  79. 79. Technologies<br />Ontology inference<br />Text<br />Text<br />Samsung releases Omnia<br />Text<br />Apple releases <br />iPhone<br />IPTV service is launched<br />Environmental<br />Technology<br />Apple supports Green technologies<br />support<br />ISA<br />beginService<br />Service<br />TV Service<br />Company<br />instanceOf<br />instanceOf<br />instanceOf<br />manufacture<br />use<br />instanceOf<br />Samsung<br />Apple<br />Product<br />instanceOf<br />beginService<br />Green<br />Technology<br />IPTV<br />support<br />manufacture<br />manufacture<br />ISA<br />ISA<br />HDTV<br />Device<br />Omnia<br />Software<br />ISA<br />instanceOf<br />iPhone<br />Smartphone<br />instanceOf<br />
  80. 80. Technologies<br />Ontology construction from Wikipedia Infobox<br />class<br />instance<br />properties<br />university<br />
  81. 81. Technologies<br />Ontology construction from text<br />2. Taxonomy Construction<br />is-a<br />3. Relation Addition<br />not is-a<br />1. Term extraction and conceptualization<br />The other<br />Final Ontology<br />Existing Ontology<br />Part-of<br />equipment-of<br />4. Integration<br />5. Verification<br />Part-of<br />equipment-of<br />
  82. 82. Technologies<br />COAT (CoreOnto Annotation Toolkit)<br /><ul><li>Term and relation annotation</li></li></ul><li>Technologies<br />Ontology construction cost reduction<br />Improve Ontology extension tech. and automation<br />Web-scale annotation by ontology extension tech.<br />2<br />Ontology extension cost reduction by automation<br />1<br />Devise ontology extension tech.<br />Cost reduction<br /><ul><li>Manual annotation cost reduction by using COAT
  83. 83. Further reduction could be possible if we can automate the process</li></ul>COAT<br />Auto<br />COAT<br />Before COAT<br />
  84. 84. CAT2ISA (cdh4696@world.kaist)<br /><ul><li>Technology for expanding semantic infrastructure
  85. 85. Extract semantic information from anonymous category system</li></li></ul><li>CAT2ISA<br /><ul><li>Extract isa/instanceOf relation
  86. 86. A instanceOf B: A is a member of set B
  87. 87. A is called 'instance', B is called 'concept'
  88. 88. A and B must share 'essential properties': Properties that makes something as itself
  89. 89. Example: <Key-Sun Choi, instanceOf, Professor>: X<Key-Sun Choi, instanceOf, Human>: O
  90. 90. B isa C: B is a subset of C</li></ul> <br /><ul><li>isa/instanceOf relation: vital component in many semantic applications(e.g. semantic search, Q&A system, etc.)</li></ul> <br /> <br />
  91. 91. Table Generator (cdh)<br /><ul><li>Summarize a text into table format based on its semantic tag</li></li></ul><li>Table Generator<br /><ul><li>Information extraction using "Ontology"
  92. 92. Ontology: Formal representation of a set of concepts within a domain and the relationships between those concepts
  93. 93. Ontology-based information extraction:</li></li></ul><li>Remaining for your home: references<br /><ul><li>History of Word Wide Web
  94. 94. Berners-Lee, Tim; Fischetti, Mark (1999). Weaving the Web. HarperSanFrancisco.
  95. 95. The Semantic Web
  96. 96. Berners-Lee, Tim; James Hendler and Ora Lassila (May 17, 2001). "The Semantic Web". Scientific American Magazine.
  97. 97. Grigoris Antoniou, Frank van Harmelen (March 31, 2008). A Semantic Web Primer, 2nd Edition
  98. 98. Ontology
  99. 99. Dean Allemang, James Hendler (May 9, 2008). Semantic Web for the Working Ontologist: Effective Modeling in RDFS and OWL. Morgan Kaufmann</li></li></ul><li>Remaining for your home: Use experiences<br /><ul><li>[1] P. Mistry, P. Maes. Quickies: Intelligent Sticky Notes. In the Proceedings of 4th International Conference on Intelligent Environments (IE08). Seattle, USA. 2008
  100. 100. [2] Max Van Kleek, Michael Bernstein, Katrina Panovich, Greg Vargas, David Karger, and mc schraefel, Note-to-Self: Examining Personal Information Keeping in a Lightweight Note-Taking Tool.. CHI, 2009
  101. 101. [3] The Tabulator, http://www.w3.org/2005/ajar/tab
  102. 102. Read [1,2,3] and use the system [2,3]
  103. 103. Try also the following system
  104. 104. http://www.evernote.com/
  105. 105. Smartphone version is available</li></li></ul><li>2nd day<br /><ul><li>Deep story about semantic technology (20 min)
  106. 106. Wikipedia DbpediaOntocloud (kekeeo@world.kaist.ac.kr)
  107. 107. What are the upside?
  108. 108. Email 3.0
  109. 109. Information Zooming
  110. 110. Mobile hyperlink
  111. 111. Personal Preference Ontology and its use
  112. 112. Collective semantic intelligence of LOD + ontology cloud
  113. 113. Another demo (5 min)
  114. 114. What you can do immediately (review)
  115. 115. What you can contribute (review)
  116. 116. Big picture
  117. 117. Function, society
  118. 118. Technology to study </li></li></ul><li>IT-Life Ontology<br />
  119. 119. IT Campus Domain Ontology (Partial)<br />
  120. 120. Wikipedia (http://en.wikipedia.org)<br /><ul><li>What is Wikipedia?
  121. 121. An online, collaboratively edited encyclopedia
  122. 122. Articles are available in over 250 languages
  123. 123. Freely available and freely distributable
  124. 124. Inter-language (interwiki) page links</li></li></ul><li>DBpedia (http://dbpedia.org)<br /><ul><li>What is the DBpedia?
  125. 125. A community effort to extract structured information from Wikipedia
  126. 126. Available on the Web
  127. 127. Different types of structured information 
  128. 128. Infobox templates: summaries of the most relevant facts contained in an article
  129. 129. Categorization information
  130. 130. Images
  131. 131. Geo-coordinates
  132. 132. Links to external Web pages </li></li></ul><li>OntoCloud<br /><ul><li>Our own constructed Ontology
  133. 133. Goals 
  134. 134. Making more intelligent IT systems focusing on devices and resources
  135. 135. Key classes
  136. 136. Device, Product, Resource, Technology, Person and Company</li></li></ul><li>Structure of OntoCloud<br /><ul><li> Template Ontology
  137. 137. Constructing the Pivot dataset
  138. 138. The infobox dataset from DBpedia3.4 (semi-automated)
  139. 139. IT CUO (IT Core Upper Ontology)
  140. 140. A middle level ontology for integration
  141. 141. Ontologies under IT domains
  142. 142. IT Service Ontology
  143. 143. IT Device Ontology
  144. 144. IT Core Ontology </li></li></ul><li>Mobile 3.0 and its Requirements (full picture: jha)<br /><ul><li>Email 3.0
  145. 145. Information Zooming
  146. 146. Mobile hyperlink
  147. 147. Personal Preference Ontology and its use
  148. 148. Collective semantic intelligence of LOD + ontology cloud</li></li></ul><li>E-mail 3.0(email categorization)<br />Automatically map into a class in ontology<br />Related Problems<br /><ul><li>Topic detection</li></ul>Current Status<br /><ul><li>Categorization of long and well-formed text (e.g. Wikipedia documents)</li></ul>Challenges<br /><ul><li>Short message interpretation
  149. 149. Personal writing styles</li></li></ul><li>E-mail 3.0(Recipient recommendation)<br />Automatically recommend person to whom the message should be sent<br />sender@abc.com<br />Challenges<br /><ul><li>Task Ontology modeling</li></li></ul><li>E-mail 3.0(Relevant information attachment)<br />Automatically attach pictures<br />sender@abc.com<br />Automatically attach files in local disk<br />Challenges<br /><ul><li>Semantic tags on multimedia data
  150. 150. Local file indexing</li></ul>The Samsung Group is composed of numerous international affiliated businesses, most of them united under the Samsung brand including Samsung Electronics, the world's largest electronics company, <br />
  151. 151. Topic<br />Information retrieval<br />Deadline<br />Mar 30, 2010<br />Organizer<br />Benno Stein<br />E-mail 3.0(Mash-up Services)<br />A message in inbox<br />Automatically create to-do list<br />The following list organizes classic and ongoing topics from the field<br />of text-based IR for which contributions are welcome:<br />- Theory. Retrieval models, language models, similarity measures,<br />formal analysis<br />- Mining and Classification. Category formation, clustering, entity<br />resolution, document classification <br />---------------------------------------------------------------------------<br />Important Dates:<br />---------------------------------------------------------------------------<br />Mar 30, 2010 Deadline for paper submission<br />Apr 20, 2010 Notification to authors<br />May 17, 2010 Camera-ready copy due<br />Aug 30, 2010 Workshop opens<br />---------------------------------------------------------------------------<br />Workshop Organization:<br />---------------------------------------------------------------------------<br />Benno Stein, Bauhaus University Weimar<br />Michael Granitzer, Know-Center Graz & Graz University of Technology<br />Contact: tir@webis.de<br />Information about the workshop can be found at http://www.tir.webis.de<br />Related Problems:<br /><ul><li>Table information extraction
  152. 152. Mash-up</li></ul>Current Status<br /><ul><li>Table information generation from text</li></ul>Challenges<br /><ul><li>Table information generation from semi-structured text</li></li></ul><li>Information Zooming<br /><ul><li> What is information zooming?
  153. 153. Show small amount of information first
  154. 154. When user requires more information about one part, shows more detailed information about that part.
  155. 155. Why is it necessary?
  156. 156. Mobile environment: small display
  157. 157. We cannot show all the necessary information at once! (Lack of space)</li></li></ul><li>Information Zooming in Semantic Post-It<br /><ul><li> Information zooming for one word</li></ul> <br /> <br /> <br /> <br /> <br /> <br /><ul><li>Information zooming for whole memo</li></li></ul><li>Mobile hyperlink<br /><ul><li> What is mobile hyperlink?
  158. 158. Represent URL as barcode
  159. 159. Take a picture of the barcode using camera in cellphone and you move to that URL!
  160. 160.  Why is it necessary?
  161. 161. Mobile environment: small interface
  162. 162. Hard to type all the URL
  163. 163. Example of mobile hyperlink</li></ul> <br /> <br /><ul><li>QR code:</li></ul>                            <br />
  164. 164. Personal Preference Ontology and its use<br /><ul><li>Task of packaging from a potentially large ontology, one or several significant sub-parts
  165. 165. Knowledge sharing and re-use crucial research issues</li></ul> <br /><ul><li>On-demand Extraction Service
  166. 166. Takes a concept and extract the relations
  167. 167.  
  168. 168. Interactive Service
  169. 169. The user have to select class and relations to consider</li></li></ul><li>Collective semantic intelligence of LOD + ontology cloud<br />
  170. 170. The Linked Open Data Cloud<br />
  171. 171. What you can do immediately<br />review<br />discussion<br />
  172. 172. What you can contribute<br /><ul><li>Data Synchronization for Mobile applications
  173. 173. Synchronization is a data transfer between computer and mobile device that aims to keep both of components in a coherent state
  174. 174. Knowledge-driven Security Handling for Mobile Applications
  175. 175. Several mobile applications attacks have beenrecently reported
  176. 176. Device  and environment
  177. 177. Ontology Packaging for Mobile field
  178. 178. Bacause of its physical aspect, a mobile device has a limited processing and computing capabilities </li></li></ul><li>Big picture<br /><ul><li>Function, society and Technology to study </li></li></ul><li>Windows Mobile<br />isDevelopedBy<br />Microsoft<br />Windows Mobile<br />isDevelopedBy<br />Microsoft<br />Omnia 2<br />ISA<br />smartphone<br />Omnia 2<br />ISA<br />smartphone<br />Flash memory <br />ISA<br />computer storage <br />Flash memory <br />ISA<br />computer storage <br />Omnia 2<br />hasOS<br />Windows Mobile<br />Omnia 2<br />hasOS<br />Windows Mobile<br />Omnia 2<br />hasMemory<br />Flash memory<br />Omnia 2<br />hasMemory<br />Flash memory<br />Big Picture<br />A working flow of Semantic Post-It<br />Contents Space<br />Omnia 2 is a multimedia smartphone announced at Samsung. Omnia 2 runs Windows Mobile and comes with flash memory..<br />Flash memory is a non-volatile computer storage that <br />can be electrically erased and reprogrammed.<br />Windows Mobile is a compact mobile operating system developed by Microsoft<br />Message Space<br />Omnia 2 is a multimedia smartphone announced at Samsung. Omnia 2 runs Windows Mobile and comes with flash memory..<br />Triple Message Space<br />(Table information)<br />Linked Triple Message Space<br />hasMemory<br />hasOS<br />What is the next step?<br />
  179. 179. Windows Mobile<br />isDevelopedBy<br />Microsoft<br />Omnia 2<br />ISA<br />smartphone<br />Flash memory <br />ISA<br />computer storage <br />Omnia 2<br />hasOS<br />Windows Mobile<br />Omnia 2<br />hasMemory<br />Flash memory<br />Big Picture<br />Message generation<br />Linked Triple Message Space<br />hasMemory<br />hasOS<br />Personalized Message Space<br />Omnia 2 is a multimedia smartphone announced at Samsung. Omnia 2 runs Windows Mobile developed by Microsoft and comes with flash memory which is a computer storage.<br />How to do so? <br />Do you have an idea how to utilize personalized ontology to generate sentences?<br />Personalized ontology<br />
  180. 180. Established<br />1971<br />Province<br />Daejeon<br />Country<br />South Korea<br />…<br />…<br />Big Picture<br />Functions (1/2)<br /><ul><li>From text to presentation file
  181. 181. Challenges
  182. 182. Semantic Tagging to Image
  183. 183. Refer to http://www.image-net.org/</li></ul>KAIST is located in Daejeon, South Korea. KAIST was established by Korean government in 1971 <br />KAIST<br />established<br />1971<br />province<br />Daejeon<br />Country<br />South Korea<br />Table information<br />Table information + images<br />
  184. 184. Big Picture<br />Functions (2/2)<br /><ul><li>From table to text
  185. 185. Generate NL text by traversing table
  186. 186. KAIST-Province-Daejeon
  187. 187. Daejeon-Districts-fifth
  188. 188.  KAIST is located in Daejeon. Daejeon is the fifth largest city in the country.
  189. 189. Challenges
  190. 190. Transform a predicate into verb phrases
  191. 191. Ex) Province -> is located in</li></li></ul><li>Big Picture<br />Society<br />Message Interpretation Services<br />HTTP request<br />Semantic Post-IT Server<br />(HTTP server)<br />Semantic Post-IT client<br />TABLEGEN<br />CAT2ISA<br />HTTP response<br />Personal Ontology<br />Ontology Access<br />DBpedia Access<br />Local Message DB<br />OpenAPI generator<br />Make your own message interpretation modules and upload it.<br />OpenAPI generator will make it available as an OpenAPI service.<br />
  192. 192. Big Picture<br />Technologies to study(interdisciplinary)<br />Architecture<br />/Urban design<br />Software Engineering<br />Graphics<br />IR, AI, MachineLearning<br />HCI<br />CognitiveScience<br />Internet of Things<br />Cloud Computing<br />Design<br />DB & <br />Data Mining<br />ConvergenceNetworks<br />Sociology<br />Middleware<br />Handle huge amount of messages<br />Ex) manipulating Wikipedia documents<br />Find person of my interests<br />Ex) References in papers<br />Which layout is suitable for the display?<br />Ex) Table-memo for a tiny display<br />Plug-in architecture<br />Ex) Collect personal documents by using Google Desktop APIs<br />How to extract table data from memo?<br />Ex) information extraction from document<br />Which type of memo? Writing style anaylsys<br />Ex) To-do list, contact, documents<br />Write message anywhere and anytime<br />Ex) RFID-equipped notes<br />
  193. 193. Deep story about semantic technology<br />discussion!<br />
  194. 194. Credits<br />Dong-Hyun Choi, cdh4696@world.kaist.ac.kr<br />Eun-Kyung Kim, kekeeo@world.kaist.ac.kr<br />JinhyunAhn, jhahn@world.kaist.ac.kr<br />Key-Sun Choi, kschoi@kaist.edu<br />http://swrc.kaist.ac.kr/ontocloud<br />http://swrc.kaist.ac.kr/SemanticToolkits/<br />
  1. A particular slide catching your eye?

    Clipping is a handy way to collect important slides you want to go back to later.

×