Recent Changes - Search:

MLFL Home

University of Massachusetts

MLFL Wiki

Mining Query Associations To Predict Search Performance

Abstract: Understanding the characteristics of queries where a search engine is failing is important in improving engine effectiveness. Previous work largely relies on user-interaction features (e.g., clickthrough statistics) to identify such underperforming queries. However, relying on interaction behavior means that searchers need to become dissatisfied and need to exhibit that in their search behavior, by which point it may be too late to help them. To overcome this, we propose a method to generate underperforming query identification rules instantly using topical and lexical attributes. The method first generates query attributes using sources such as topics, concepts (entities), and keywords appearing in queries. Then, association rules are learned by exploiting the FP-growth algorithm and decision trees using underperforming query examples. We develop a query classification model capable of accurately estimating dissatisfaction using the generated rules, and demonstrate significant performance gains over state-of-the-art query performance prediction models.

Bio: Youngho Kim is a Ph.D. student in Department of Computer Science at the University of Massachusetts, Amherst, in which he started his doctoral study in 2009. His doctoral studies are supervised by Professor W. Bruce Croft. He received a B.Sc. degree in 2006 from the Inha University, South Korea, and an M.Sc. degree in 2009 from the Korea Advanced Institute of Science and Technology (KAIST), South Korea. His research interests span the fields of information retrieval and natural language processing, including retrieval models, query processing, user analysis, text-data mining and sentiment analysis. He published 10 papers related to these fields.

Edit - History - Print - Recent Changes - Search
Page last modified on October 27, 2012, at 04:47 PM