Abstract : This paper focuses on building accurate profiles of users, based on bookmarking systems. To achieve this goal, we define personalized parsimonious language models that employ three main resources: the tags, the documents tagged by the user and word embeddings that handle general knowledge. Experiments completed on Delicious data show that our proposal outperforms state-of-the-art approaches and non-personalized parsimonious models.
https://hal-ujm.archives-ouvertes.fr/ujm-01615362
Contributor : Mathias Géry
<>
Submitted on : Thursday, October 12, 2017 - 12:08:13 PM
Last modification on : Tuesday, February 12, 2019 - 1:31:00 AM
Nawal Ould Amer, Philippe Mulhem, Mathias Géry. Personalized Parsimonious Language Models for User Modeling in Social Bookmaking Systems. European Conference on Information Retrieval, Apr 2017, Aberdeen, United Kingdom. ⟨ujm-01615362⟩
