Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
.nju.edu.cn                      BipRank: Ranking and Summarizing                         RDF Vocabulary Descriptions     ...
Outline                                    ws .nju.edu.cn        Introduction        Salience measurement        Vocabular...
Vocabularies and Linked Data                                                  ws .nju.edu.cn   Vocabularies               ...
Vocabulary search engines                                    ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn   4 of 25
Vocabularies                                    ws .nju.edu.cn                                      ScaleGong Cheng (程龚) g...
Vocabulary snippets --- state of the art                                               ws .nju.edu.cnGong Cheng (程龚) gchen...
Vocabulary snippets --- our approach                                           ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.ed...
Vocabulary summarization                                                                            ws .nju.edu.cn        ...
Outline                                    ws .nju.edu.cn        Introduction        Salience measurement        Vocabular...
A bipartite view of vocabulary description                                                 ws .nju.edu.cnGong Cheng (程龚) g...
Surfer behavior --- type A                                    ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn   11 of 25
Surfer behavior --- type B                                    ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn   12 of 25
BipRank                                                                     ws .nju.edu.cn       Next step                ...
Pattern of RDF sentence                                    ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn   14 of 25
p(s|u)                                                                         ws .nju.edu.cn        Frequency of Pattern(...
Evaluation setting                                                                           ws .nju.edu.cn        Test ca...
Evaluation results                                    ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn   17 of 25
Outline                                    ws .nju.edu.cn        Introduction        Salience measurement        Vocabular...
Goodness of a summary                                                           ws .nju.edu.cn        Salience        Quer...
Looking for the best summary                                              ws .nju.edu.cn        Multi-objective optimizati...
Evaluation setting                                                                      ws .nju.edu.cn        Judges      ...
Evaluation results                                    ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn   22 of 25
Performance testing                                                                ws .nju.edu.cn                         ...
Outline                                    ws .nju.edu.cn        Introduction        Salience measurement        Vocabular...
Conclusions                                                           ws .nju.edu.cn        Salience measurement          ...
Upcoming SlideShare
Loading in …5
×

BipRank: Ranking and Summarizing RDF Vocabulary Descriptions

820 views

Published on

Published in: Technology, Education
  • Be the first to comment

BipRank: Ranking and Summarizing RDF Vocabulary Descriptions

  1. 1. .nju.edu.cn BipRank: Ranking and Summarizing RDF Vocabulary Descriptions Gong Cheng1, Feng Ji2, Shengmei Luo2, Weiyi Ge1, Yuzhong Qu11State Key Laboratory for Novel Software Technology, Nanjing University, China 2Communication Services R&D Institute, ZTE Corporation, China Presented at JIST2011
  2. 2. Outline ws .nju.edu.cn Introduction Salience measurement Vocabulary summarization ConclusionsGong Cheng (程龚) gcheng@nju.edu.cn 2 of 25
  3. 3. Vocabularies and Linked Data ws .nju.edu.cn Vocabularies Your own vocabulary Reuse Linked DataGong Cheng (程龚) gcheng@nju.edu.cn 3 of 25
  4. 4. Vocabulary search engines ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn 4 of 25
  5. 5. Vocabularies ws .nju.edu.cn ScaleGong Cheng (程龚) gcheng@nju.edu.cn 5 of 25
  6. 6. Vocabulary snippets --- state of the art ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn 6 of 25
  7. 7. Vocabulary snippets --- our approach ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn 7 of 25
  8. 8. Vocabulary summarization ws .nju.edu.cn Vocabulary summarization = ranking and selecting RDF sentencesGong Cheng (程龚) gcheng@nju.edu.cn 8 of 25
  9. 9. Outline ws .nju.edu.cn Introduction Salience measurement Vocabulary summarization ConclusionsGong Cheng (程龚) gcheng@nju.edu.cn 9 of 25
  10. 10. A bipartite view of vocabulary description ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn 10 of 25
  11. 11. Surfer behavior --- type A ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn 11 of 25
  12. 12. Surfer behavior --- type B ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn 12 of 25
  13. 13. BipRank ws .nju.edu.cn Next step ? Uniform Current step type-A behavior type-B behaviorGong Cheng (程龚) gcheng@nju.edu.cn 13 of 25
  14. 14. Pattern of RDF sentence ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn 14 of 25
  15. 15. p(s|u) ws .nju.edu.cn Frequency of Pattern(s) #RDF_sentence in the vocabulary that has the same pattern Popularity of Pattern(s) #Vocabulary in the repository that has the same patternGong Cheng (程龚) gcheng@nju.edu.cn 15 of 25
  16. 16. Evaluation setting ws .nju.edu.cn Test cases 9 moderate-sized vocabularies randomly selected from Falcons Gold standard Salience given by 6 human experts Competitors Cp: Zhang et al. (WWW2007) Our approach BipRank-U: pattern-unaware BipRank-F: using pattern frequency BipRank-P: using pattern popularity Metric Pearson product-moment correlation coefficientGong Cheng (程龚) gcheng@nju.edu.cn 16 of 25
  17. 17. Evaluation results ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn 17 of 25
  18. 18. Outline ws .nju.edu.cn Introduction Salience measurement Vocabulary summarization ConclusionsGong Cheng (程龚) gcheng@nju.edu.cn 18 of 25
  19. 19. Goodness of a summary ws .nju.edu.cn Salience Query relevance Textual similarity between query and summary Cohesion Term overlap between RDF sentencesGong Cheng (程龚) gcheng@nju.edu.cn 19 of 25
  20. 20. Looking for the best summary ws .nju.edu.cn Multi-objective optimization Single aggregate objective function Solution: a greedy strategyGong Cheng (程龚) gcheng@nju.edu.cn 20 of 25
  21. 21. Evaluation setting ws .nju.edu.cn Judges 18 human experts Test cases 190 searches over 2,012 vocabularies crawled by Falcons Competitors Generic: Zhang et al. (WWW2007) Our approach QR: query relevance QR+S: query relevance + salience QR+C: query relevance + cohesion Metric Rating on a 10-point scaleGong Cheng (程龚) gcheng@nju.edu.cn 21 of 25
  22. 22. Evaluation results ws .nju.edu.cnGong Cheng (程龚) gcheng@nju.edu.cn 22 of 25
  23. 23. Performance testing ws .nju.edu.cn Size of summary Runtime Size of vocabularyGong Cheng (程龚) gcheng@nju.edu.cn 23 of 25
  24. 24. Outline ws .nju.edu.cn Introduction Salience measurement Vocabulary summarization ConclusionsGong Cheng (程龚) gcheng@nju.edu.cn 24 of 25
  25. 25. Conclusions ws .nju.edu.cn Salience measurement Sentence-term graph BipRank Pattern of RDF sentence Vocabulary summarization Salience Query relevance Cohesion Implemented in Falcons Ontology Search http://ws.nju.edu.cn/falcons/ontologysearch/Gong Cheng (程龚) gcheng@nju.edu.cn 25 of 25

×