Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics

1,606 views
1,540 views

Published on

E

Published in: Technology, Travel

Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics

  1. 1. EC-WEB 2013 - 14th International Conference on Electronic Commerce and WebTechnologies Prague (Czech Republich) - 28.08.13 Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics Cataldo Musto, Giovanni Semeraro, Pasquale Lops, Marco de Gemmis
  2. 2. scenario. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  3. 3. Prague, EC-Web conference C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  4. 4. dinner time C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  5. 5. me and Pasquale like sushi. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  6. 6. what doesTripAdvisor suggest? C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  7. 7. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  8. 8. Good Price C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  9. 9. Good Reviews C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  10. 10. ....Strasbourg? C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  11. 11. 612 km far away. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  12. 12. 612 km far away. Too much :-) C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  13. 13. we ate pizza, as usual. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  14. 14. why did theTripAdvisor recommendation engine fail? C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  15. 15. it doesn’t take into account contextual information. context plays a key role in several recommendation tasks. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  16. 16. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 I attended last Sigur Ròs concert in Rome, so I like them
  17. 17. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 I attended last Sigur Ròs concert in Rome, so I like them but their ambient music is not good if I need music recommendation for my running session!
  18. 18. a real-world recommendation engine needs to take into account contextual information C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  19. 19. what is context? • Over 150 definitions, spread across several domains (Bazire and Brezillon, 2005) • Recommender Systems area • “A set of factors that have influence on user perception and acceptance of a particular item” • A fixed set of dimensions with appropriate attributes C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  20. 20. what is context? C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 task
  21. 21. what is context? C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 task
  22. 22. what is context? C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 task company
  23. 23. what is context? C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 task company
  24. 24. what is context? C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 task moodcompany
  25. 25. what is context? C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 task company mood
  26. 26. a real-world recommendation engine needs to take into account contextual information C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  27. 27. Our contribution contextual eVSM a context-aware content-based recommendation framework based on distributional semantics C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  28. 28. timeline 2013: contextual eVSM 2010-2012: eVSM(*) a content-based recommendation framework based on distributional semantics context-aware extension of eVSM (*) Cataldo Musto: Enhanced vector space models for content-based recommender systems. RecSys 2010: 361-364 C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  29. 29. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 eVSM a brief recap
  30. 30. eVSM • Content-based Recommendation Framework • (semantic) vector space representation based on distributional models • negative information modeled through quantum negation operator • recommendation seen as a form of similarity in vector spaces • four different profiling models • cosine similarity to get the top-k recommendations C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 cornerstones
  31. 31. distributional models (Firth, 1957) Firth, J.R. A synopsis of linguistic theory 1930-1955. In Studies in Linguistic Analysis, pp. 1-32, 1957. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 cornerstone 1
  32. 32. “meaning is its use” L.Wittgenstein (Austrian philosopher) distributional semantics C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  33. 33. insight by analyzing large corpus of textual data it is possible to infer information about the usage (about the meaning) of the terms. example distributional models C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  34. 34. distributional hypothesis C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 “terms that occur in similar contexts share a similar meaning”
  35. 35. distributional models c1 c2 c3 c4 c5 c6 rock ✔ ✔ ✔ post rock ✔ ✔ jazz ✔ classical ✔ ✔ ✔ term/context matrix (WordSpace) C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  36. 36. distributional models rock vs. post rock = good overlap c1 c2 c3 c4 c5 c6 rock ✔ ✔ ✔ post rock ✔ ✔ jazz ✔ classical ✔ ✔ ✔ C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  37. 37. distributional models rock vs. classical = no overlap c1 c2 c3 c4 c5 c6 rock ✔ ✔ ✔ post rock ✔ ✔ jazz ✔ classical ✔ ✔ ✔ C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  38. 38. • Key: definition of what is the ‘context’ • Different granularities are possible • Document • Paragraph • Sentence • Sliding window of words distributional models C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  39. 39. rock alternative-rock post-rock classical example WordSpace C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  40. 40. representation of documents (*) can be inferred by combining the representation of the terms (**) occurring in the document. (*) documents = artists (**) terms = tags C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  41. 41. distributional models c1 c2 c3 c4 c5 c6 c7 c8 c9 t2 ✔ ✔ ✔ ✔ t3 ✔ ✔ ✔ d1 ✔ ✔ ✔ ✔ ✔ term/context matrix (DocSpace) C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  42. 42. Coldplay Radiohead Kings of Leon Lady Gaga example DocSpace C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  43. 43. distributional models • Features • semantic vector space representation of terms and documents (user profiles and items !) • light semantics, based on term co-occurrences in large corpus of data • based on distributional hypothesis • totally unsupervised • just based on the analysis of term distribution C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  44. 44. quantum negation (Widdows, 2007) cornerstone 2 C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  45. 45. negation inVSMs •Widdows proposed a different point of view • Negation view as a form of orthogonality between vectors • Vision inherited from Quantum Logic state of the art C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  46. 46. negation inVSMs • Some theory • Given vector a and vector b • Through quantum negation it is possible to define a vector a not b (a ∧¬b) • Projection of vector a on the subspace orthogonal to those generated by vector b Quantum Negation C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  47. 47. negation inVSMs • Two terms can be considered as mutually unrelevant if they never co-occur • Two documents can be considered as mutually unrelevant if they have no features in common • No features in common scalar product = 0 •Orthogonality Widdows’ insights C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  48. 48. quantum negation • Vector A models positive feedbacks • Information about what a user likes • Vector B models negative feedbacks • Information about what a user does not like • Vector A not B combines both information sources application to CBRS C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  49. 49. recommendation step C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 cornerstone 3
  50. 50. ThresholdRatingItems Higher weight given to the documents with higher rating User Profiles Weighted Random Indexing-based (w-RI) C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  51. 51. VSM representation of wQN-based profile for user u Positive User ProfileVector Negative User ProfileVector User Profiles Weighted Quantum Negation-based (w-QN) C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  52. 52. scenario football news sports news politics news politics news user profile Recommendation task seen as similarity calculation between vectors C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  53. 53. scenario football news sports news politics news politics news user profile football and sports news are recommended to the target user C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  54. 54. size=400 - Movielens dataset Gap always around 1% 84 84,75 85,5 86,25 87 p@1 P@3 P@5 P@10 84,7584,7 84,97 85,39 84,58 84,7 84,85 85,27 84,5 84,77 84,4384,47 85,5285,58 86,0185,94 eVSM VSM LSI Bayes experiment(*) Cataldo Musto - Enhanced Vector Space Models for Content-based Recommender Systems - Ph.D. defense - University of Bari Aldo Moro, Italy - 08.06.12 (*) Cataldo Musto. Enhanced Vector Space Models for content-based Recommender Systems. Ph.D dissertation, 2012
  55. 55. 84 84,75 85,5 86,25 87 p@1 P@3 P@5 P@10 84,7584,7 84,97 85,39 84,58 84,7 84,85 85,27 84,5 84,77 84,4384,47 85,5285,58 86,0185,94 eVSM VSM LSI Bayes size=400 - Movielens dataset Significant Improvement Cataldo Musto - Enhanced Vector Space Models for Content-based Recommender Systems - Ph.D. defense - University of Bari Aldo Moro, Italy - 08.06.12 experiment(*) (*) Cataldo Musto. Enhanced Vector Space Models for content-based Recommender Systems. Ph.D dissertation, 2012
  56. 56. eVSM outperforms other state-of-the-art approaches C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  57. 57. eVSM outperforms other state-of-the-art approaches C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 but it doesn’t take into account contextual information.
  58. 58. contextual eVSM C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 how to make the eVSM context-aware?
  59. 59. context-aware RSs C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 state of the art: pre and post-filtering
  60. 60. we implemented two contextualization strategies C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 both pre and post-filtering
  61. 61. context-aware eVSM •Microprofiling •based on contextual pre- filtering •Insight: to filter data before building user profiles, building a separate user profile for each contextual dimension C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  62. 62. context-aware eVSM •Contextual eVSM •based on contextual post- filtering •Insight: to calculate un- contextualized recommendation and to re-rank them according to contextual constraints C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  63. 63. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 if a user needs suggestions for a restaurant for a romantic dinner, only her ratings expressed for previous romantic dinners have to be taken into account micro-profiling insight
  64. 64. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 micro-profiling context-aware eVSM
  65. 65. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 micro-profiling context-aware eVSM user
  66. 66. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 micro-profiling context-aware eVSM contextual dimension (e.g task) and value (e.g. running)
  67. 67. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 micro-profiling context-aware eVSM sum over all the items rated under that specific contextual constraints
  68. 68. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 micro-profiling context-aware eVSM weighted with the (normalized) rating provided by the user
  69. 69. contextual eVSM insight C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 context is just a factor that can (positively or negatively) influence user preference on a certain item
  70. 70. contextual eVSM insight C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 context is just a factor that can (positively or negatively) influence user preference on a certain item context-aware recommendation can be obtained by combining an uncontextual user profile with a vector space representation of the target context
  71. 71. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 contextual eVSM context-aware eVSM
  72. 72. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 contextual eVSM context-aware eVSM user profile is the combination of two components
  73. 73. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 contextual eVSM context-aware eVSM (1) uncontextual user profile
  74. 74. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 contextual eVSM context-aware eVSM (2) vector space representation of the context
  75. 75. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 contextual eVSM context-aware eVSM weight of each component
  76. 76. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 contextual eVSM context-aware eVSM if alpha = 1, we have uncontextual recommendation since context weight = 0 !
  77. 77. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 contextual eVSM context-aware eVSM problem: how to build it ?
  78. 78. • For each contextual dimension, it exists a set of terms that is descriptive of items relevant in that specific context C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 context representation assumptions
  79. 79. • For each contextual dimension, it exists a set of terms that is descriptive of items relevant in that specific context • e.g. candlelight or seaview are relevant features if I’m looking for a restaurant for a romantic night ! C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 context representation assumptions
  80. 80. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 context representation idea to represent the context as the combination of the terms occurring in the descriptions of the items labeled as relevant under that specific contextual situation
  81. 81. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 context representation formula PRE-WRI is exactly the combination of the terms occurring in the descriptions of the items labeled as relevant under that specific contextual situation ! we adopted PRE-WRI as vector space representation of the context
  82. 82. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 rationale contextual eVSM if the user didn’t express any preference in that specific context, the vector is null, so she will receive uncontextual recommendation
  83. 83. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 rationale contextual eVSM if the user didn’t express any preference in that specific context, the vector is null, so she will receive uncontextual recommendation otherwise, a greater weight will be given to the features that are informative and relevant in the target context, that is to say, those contained in the context vector!
  84. 84. experimental evaluation. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  85. 85. •Experiments • (1) Does contextual eVSM outperform eVSM ? • (2) How does our approach performs with respect to current literature? experimental design C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  86. 86. • Movie recommendation (IMDB data) • 202 movies (textual features crawled from Wikipedia) • 62 users and 1457 ratings • 4 contextual dimensions • TIME (weekend, weekday) • PLACE (theather, home) • COMPANION (alone, friends, boyfriend, family) • MOVIE-RELATED (release week or not) experimental design dataset C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  87. 87. • Dataset and experimental settings replicate Adomavicius’ experiment (*) • Evaluation over 9 different contextual settings • Home, Friends, Non-release,Weekend,Weekday, GBFriends,TheatherWeekend andTheatherFriends • Experimental protocol: bootstrapping • 29/30th of the data as training • 1/30th as test • Randomly generated, 500 runs experimental design settings C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 (*) G.Adomavicius et al. , Incorporating contextual information in recommender systems using a multi- dimensional approach.ACM Trans. Inf. Systems, 2005
  88. 88. • eVSM settings • Uncontextual baselines • WRI and WQN • Pre-filtering • PRE-WRI and PRE-WQN • Post-filtering: • CONTEXT-WRI and CONTEXT-WQN • alpha=0.5, alpha=0.8 • 8 settings for each run experimental design parameters C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  89. 89. experiment 1 selection of results - HOME segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 48,75 52,5 56,25 60 pre-filtering outperforms the baseline (it’s an exception, in the other segments it doesn’t !) C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  90. 90. experiment 1 selection of results - HOME segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 48,75 52,5 56,25 60 contextual eVSM improves the F1 measure C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  91. 91. experiment 1 selection of results - HOME segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 48,75 52,5 56,25 60 alpha=0.8 better than alpha 0.5 C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  92. 92. experiment 1 selection of results - HOME segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 48,75 52,5 56,25 60 contextual eVSM with negation is the best setting C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  93. 93. experiment 1 selection of results - FRIEND segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 47,5 50 52,5 55 similar outcomes C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  94. 94. experiment 1 selection of results - FRIEND segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 47,5 50 52,5 55 pre-filtering doesn’t improve the accuracy C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  95. 95. experiment 1 selection of results - FRIEND segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 47,5 50 52,5 55 contextual eVSM generally does C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  96. 96. experiment 1 selection of results - FRIEND segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 47,5 50 52,5 55 contextual eVSM with negation and alpha=0.8 is always the best setting C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  97. 97. experiment 1 selection of results - NON RELEASE segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 48 51 54 57 contextual eVSM with negation and alpha=0.8 is always the best setting C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  98. 98. experiment 1 selection of results - NON RELEASE segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 48 51 54 57 typically, alpha=0.8 is better than alpha 0.5 C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  99. 99. experiment 1 selection of results - NON RELEASE segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 48 51 54 57 outcome: context has just a little influence on user perception and acceptance. Uncontextual preferences are still the “core”. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  100. 100. experiment 1 selection of results - NON RELEASE segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 48 51 54 57 outcome: context has to be taken into account, but just a little! C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  101. 101. experiment 1 selection of results - THEATHER segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 47,25 49,5 51,75 54 However, in some setting contextual eVSM without negation is the best C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  102. 102. experiment 1 selection of results - THEATHER segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 47,25 49,5 51,75 54 experiments showed a clear relationship between the amount of negative ratings and the best configurations C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  103. 103. experiment 1 selection of results - THEATHER segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 45 47,25 49,5 51,75 54 when the dataset is well balanced, cont-WQN outperforms cont-WRI, otherwise when few negative data are available, cont-WRI is the best C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  104. 104. experiment 1 selection of results - THEATHER+FRIENDS segment WRI PRE-WRI CONT-WRI-0.5 CONT-WRI-0.8 WQN PRE-WQN CONT-WQN-0.5 CONT-WQN-0.8 40 41,75 43,5 45,25 47 contextual eVSM didn’t improve F1 measure in just 1 contextual segment out of 9 C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  105. 105. experiment 1 - outcome contextual eVSM improves the predictive accuracy of eVSM C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  106. 106. experiment 2 comparison with state-of-the-art HOME FRIENDS WEEKEND THEATHER NONRELEASE WEEKDAY GBFRIEND THEAT-WEEK THEAT-FRIENDS 37 45,25 53,5 61,75 70 comparison with Adomavicius’ approach based on CF C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  107. 107. experiment 2 comparison with state-of-the-art HOME FRIENDS WEEKEND THEATHER NONRELEASE WEEKDAY GBFRIEND THEAT-WEEK THEAT-FRIENDS 37 45,25 53,5 61,75 70 eVSM outperforms CF in 6 segments out of 9 C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  108. 108. experiment 2 - outcome contextual eVSM improves the predictive accuracy of state-of-the-art approaches based on CF C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  109. 109. recap. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  110. 110. recap • context-aware eVSM • baseline: eVSM • content-based recommendation framework • distributional semantics + quantum negation • evolution: contextual eVSM • comparison of two approaches for context-aware eVSM • exact pre-filtering and weighted post-filtering • large experimental evaluation, comparison with state of the art approaches C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013 C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  111. 111. contextual eVSM overcomes state-of-the-art approaches C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  112. 112. future research. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  113. 113. evaluation with different datasets C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  114. 114. open knowledge sources and linked data for CBRS. C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  115. 115. evaluation with user-based metrics (serendipity, novelty, unexpectedness) C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013
  116. 116. questions? Cataldo Musto, Ph.D cataldo.musto@uniba.it C. Musto, G. Semeraro, P. Lops, M. de Gemmis. - Contextual eVSM: a context-aware content-based recommendation framework based on distributional semantics - EC WEB 2013 - Prague, Czech Republic - 28.08.2013

×