Skip to Main content Skip to Navigation
Journal articles

Modéliser la diversité au cours du temps pour détecter le contexte dans un service de musique en ligne

Amaury l'Huillier 1 Sylvain Castagnos 1 Anne Boyer 1
1 KIWI - Knowledge Information and Web Intelligence
LORIA - AIS - Department of Complex Systems, Artificial Intelligence & Robotics
Abstract : Many studies have shown that taking into account the context improves the quality of recommender systems. However, traditional methods infer the context using personal data (location, date, age, etc.). In this paper, we propose to automatically detect context changes, without knowledge on users (explicit context), but based on common features of consulted items (implicit context). To do this, we propose a formal model which can establish a correspondence between the variation of diversity over time whithin the paths of users and context changes. This model has been tested on a musical corpus of more than 200,000 tracks. To validate the relevance of our model, we sought to retrieve events from the detected changes of context: our model has recovered 88% of session ends.
Document type :
Journal articles
Complete list of metadata

Cited literature [60 references]  Display  Hide  Download

https://hal.inria.fr/hal-01300419
Contributor : Sylvain Castagnos <>
Submitted on : Monday, May 2, 2016 - 3:49:51 PM
Last modification on : Tuesday, December 18, 2018 - 4:40:21 PM
Long-term archiving on: : Tuesday, May 24, 2016 - 4:45:58 PM

Files

LHuillierCastagnosBoyer-TSI201...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01300419, version 1

Collections

Citation

Amaury l'Huillier, Sylvain Castagnos, Anne Boyer. Modéliser la diversité au cours du temps pour détecter le contexte dans un service de musique en ligne. Revue des Sciences et Technologies de l'Information - Série TSI : Technique et Science Informatiques, Lavoisier, 2016. ⟨hal-01300419⟩

Share

Metrics

Record views

346

Files downloads

325