A Personalized Location Aware Multi-Criteria Recommender System Based on Context-Aware User Preference Models

Abstract : Recommender Systems have been applied in a large number of domains. However, current approaches rarely consider multiple criteria or the level of mobility and location of a user. In this paper, we introduce a novel algorithm to construct personalized multi-criteria Recommender Systems. Our algorithm incorporates the user’s current context, and techniques from the Multiple Criteria Decision Analysis field of study to model user preferences. The obtained preference model is used to assess the utility of each item, to then recommend the items with the highest utility. The criteria considered when creating preference models are the user location, mobility level and user profile. The latter is obtained considering the user requirements, and generalizing the user data from a large-scale demographic database. The evaluation of our algorithm shows that our system accurately identifies the demographic groups where a user may belong, and generates highly accurate recommendations that match his/her preference value scale.
Document type :
Conference papers
Complete list of metadatas

Cited literature [9 references]  Display  Hide  Download

https://hal.inria.fr/hal-01459641
Contributor : Hal Ifip <>
Submitted on : Tuesday, February 7, 2017 - 1:06:08 PM
Last modification on : Thursday, February 7, 2019 - 2:50:01 PM
Long-term archiving on: Monday, May 8, 2017 - 2:00:44 PM

File

978-3-642-41142-7_4_Chapter.pd...
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

Salvador Valencia Rodríguez, Herna Viktor. A Personalized Location Aware Multi-Criteria Recommender System Based on Context-Aware User Preference Models. 9th Artificial Intelligence Applications and Innovations (AIAI), Sep 2013, Paphos, Greece. pp.30-39, ⟨10.1007/978-3-642-41142-7_4⟩. ⟨hal-01459641⟩

Share

Metrics

Record views

121

Files downloads

173