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.

Ricochet: Context and Complementarity-Aware, Ontology-based POIs Recommender System

In this paper we propose a new approach for improving the personalization of POIs recommender system. Existing context-aware POIs recommender sys- tems usually take into account only peripheral contextual variables. We present Ricochet, an ontology-based system that refines the recommendation results by implementing an inter-POI parameter that we call the “complementarity”. We show how this new parameter can generate more effective recommendations. Our experiments are grounded using data from the location-based social network (LBSN) Yelp.com.

  • Be the first to comment

Ricochet: Context and Complementarity-Aware, Ontology-based POIs Recommender System

  1. 1. Ricochet: Context and Complementarity-Aware, Ontology-based POIs Recommender System Chun Lu, Philippe Laublet, Milan Stankovic SALAD 2014 May 26th, 2014 Crete, Greece
  2. 2. Outline —  Introduction —  Ricochet system —  Evaluation —  Conclusion
  3. 3. Introduction —  Point of interest (POI) —  Increasing popularity of POIs recommender systems —  Dissatisfaction with recommendations of existing systems
  4. 4. What’s wrong?
  5. 5. What’s wrong? —  Before & after a check-in, recommendations remain unchanged. —  Research questions —  User study
  6. 6. Ricochet System Criteria —  Contextual criteria —  Intrinsic criteria —  Criteria of complementarity
  7. 7. Ricochet System Criteria of complementarity —  Each POI category is mapped to a particular feeling. [Savage et al., 2012] Arts & Entertainment = "feeling artsy" Nightlife = "feeling like a party animal" College & Education = "feeling nerdy" Great Outdoors = "feeling outdoorsy" Food = "feeling hungry" Shops = "feeling shopaholic"
  8. 8. Ricochet System Criteria of complementarity —  Having a certain feeling à Go to a certain POI —  Visit a POI à Having a certain feeling —  Just-visited POI causes a feeling —  Going-to POI satisfies it
  9. 9. Ricochet System Criteria of complementarity —  Physical activities possess a specific assessable intensity. [Tapia et al., 2007] —  Intensity of expressiveness —  Cognitive, emotional, physical —  POIs classified into four intensity levels —  Daily life = alternation of different intensities
  10. 10. Ricochet System Sample of OntoPOI
  11. 11. Ricochet System Recommendation engine —  API : Yelp & Jena & Google Maps
  12. 12. Evaluation —  Two versions of Ricochet —  Process —  Metrics: precision, recall, normalized Discounted Cumulative Gain
  13. 13. Results —  The complementarity improves the relevance of the recommendations.
  14. 14. Conclusion & Future work —  Ricochet : improve the relevance by considering the complementarity —  Refine the complementarity —  Integrate to existing systems
  15. 15. Questions & Answers Thanks for your attention! Any questions?

×