Personalized Facet Recommendation based on Conditional Random Fields

DOI

10.2991/icmmita-15.2015.41How to use a DOI?

Keywords

Facet Recommendation; Personalized Recommendation; Conditional Random Fields

Abstract

Faceted search is a kind of exploratory search, which is the complementary of keyword search. The current faceted search techniques display all facets. When there are plenty of facets, they are not available. Existing facet recommendation approaches mainly select facets according to the experts’ experience or the statistics. They do not consider the user’s interest which affects the recommendation result. Thus, this paper proposes a personalized facet recommendation approach based on conditional random fields. First, we use the user’s query logs to build user profile and regard the facets in the logs which the user selects as his/her interested ones. Second, we select multiple kinds of features and use conditional random fields to construct the facet classification model. At last, when a user submits a query, the system selects all candidate facets and uses the facet classification model to predict their interest degrees. Then it sorts these facets from the highest degree to the lowest degree and shows top-k facets to the user. Experimental results validate the effectiveness of our approach.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yongquan Dong
AU  - Qiang Chu
AU  - Ping Ling
PY  - 2015/11
DA  - 2015/11
TI  - Personalized Facet Recommendation based on Conditional Random Fields
BT  - Proceedings of the 2015 3rd International Conference on Machinery, Materials and Information Technology Applications
PB  - Atlantis Press
SP  - 207
EP  - 210
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmmita-15.2015.41
DO  - 10.2991/icmmita-15.2015.41
ID  - Dong2015/11
ER  -