Please use this identifier to cite or link to this item: http://hdl.handle.net/2080/2705
Full metadata record
DC FieldValueLanguage
dc.contributor.authorMazumdar, Pramit-
dc.contributor.authorPatra, Bidyut Kumar-
dc.contributor.authorBabu, Korra Sathya-
dc.date.accessioned2017-05-02T10:12:04Z-
dc.date.available2017-05-02T10:12:04Z-
dc.date.issued2017-03-
dc.identifier.citationInternational Conference on Management of Data (COMAD), IIT Madras, Chennai, India, 8-10 March 2017en_US
dc.identifier.urihttp://hdl.handle.net/2080/2705-
dc.descriptionCopyright for this paper belongs to proceeding pubisheren_US
dc.description.abstractThe advances in GPS enabled mobile devices has led people to share their real-time experiences ubiquitously through various online platforms. Users’ experiences on various services (products), interesting places (famously known as point of interests (POI)), events, movies, etc. are being widely collected these days by various enterprise houses. It helps the community to browse through these online contents before selecting a product or a POI. Recommender system is proven to be a successful tool which can automatically provide an effective list of items to an active user based on her preferences by filtering through the large item space. However, a recommender system often fails to learn the user preferences for a new user who has no historical data (widely known as the cold start problem). This work focuses on developing a POI recommender system for handling various cold start scenarios such as, ‘new user’ and ‘new city’. In this regard, a Feature and Region based POI Recommender System (FRRS) has been devised which can effectively provide a list of top-K POIs to an active user in cold start scenarios. The proposed system FRRS has two modules, modelling and recommendation. First, the user preferences and features of POIs are learnt from various online contents such as ratings and reviews. Finally, the recommendations are obtained by combining the learnt user preferences with the interests of influential users and the proximity of POIs from the active location for recommending a list of top-K POIs. Experiments are performed on the real-world Yelp dataset. We compare the performance of our approach with three existing works and a baseline approach for recommending POIs.The obtained results show that our proposed approach out-performs the existing works.en_US
dc.subjectPOI Recommendationen_US
dc.subjectColdstart Scenariosen_US
dc.titleAn Effective POI Recommendation in Various Coldstart Scenariosen_US
dc.typeArticleen_US
Appears in Collections:Conference Papers

Files in This Item:
File Description SizeFormat 
2017_COMAD_BKPatra.pdf358.29 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.