Making Collaborative Group Recommendations Based on Modal Symbolic Data

Sergio Queiroz 1 Francisco de Carvalho
1 DECISION
LIP6 - Laboratoire d'Informatique de Paris 6
Abstract : In recent years, recommender systems have achieved great success. Popular sites give thousands of recommendations every day. However, despite the fact that many activities are carried out in groups, like going to the theater with friends, these systems are focused on recommending items for sole users. This brings out the need of systems capable of performing recommendations for groups of people, a domain that has received little attention in the literature. In this article we introduce a novel method of making collaborative recommendations for groups, based on models built using techniques from symbolic data analysis. After, we empirically evaluate the proposed method to see its behaviour for groups of different sizes and degrees of homogeneity, and compare the achieved results with both an aggregation-based methodology previously proposed and a baseline methodology. The authors would like to thank CNPq and CAPES (Brazilian Agencies) for their financial support.
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01503223
Contributor : Lip6 Publications <>
Submitted on : Thursday, April 6, 2017 - 5:22:52 PM
Last modification on : Thursday, March 21, 2019 - 1:06:17 PM

Links full text

Identifiers

Citation

Sergio Queiroz, Francisco de Carvalho. Making Collaborative Group Recommendations Based on Modal Symbolic Data. SBIA 2004 - 17th Brazilian Symposium on Artificial Intelligence, Sep 2004, Sao Luis, Maranhao, Brazil. pp.307-316, ⟨10.1007/978-3-540-28645-5_31⟩. ⟨hal-01503223⟩

Share

Metrics

Record views

62