Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

1-2012

Abstract

Query suggestion is an assistive technology mechanism commonly used in search engines to enable a user to formulate their search queries by predicting or completing the next few query words that the user is likely to type. In most implementations, the suggestions are mined from query log and use some simple measure of query similarity such as query frequency or lexicographical matching. In this paper, we propose an alternative method of presenting query suggestions by their thematic topics. Our method adopts a document-centric approach to mine topics in the corpus, and does not require the availability of a query log. The heart of our algorithm is a probabilistic topic model that assumes that topics are multinomial distributions of words, and jointly learns the co-occurrence of textual words and the visual information in the video stream. Empirical results show that this alternate way of organizing query suggestions can better elucidate the high level query intent, and more effectively help a user meet his information need.

Keywords

Topic Modeling, Latent Dirichlet Allocation, Query Suggestion

Discipline

Databases and Information Systems | Graphics and Human Computer Interfaces

Research Areas

Data Science and Engineering

Publication

MultiMedia Modeling: 18th International Conference, MMM 2012, Klagenfurt, Austria, January 4-6: Proceedings

Volume

7131

First Page

288

Last Page

299

ISBN

9783642273544

Identifier

10.1007/978-3-642-27355-1_28

Publisher

Springer

City or Country

Cham

Additional URL

https://doi.org/10.1007/978-3-642-27355-1_28

Download

Share

COinS