Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

11-2007

Abstract

Searching answers to complex questions is a challenging IR task. In this paper, we examine the use of query templates with semantic slots to formulate slot-based queries. These queries have query terms assigned to entity and relationship slots. We develop several query expansion methods for slot-based queries so as to improve their retrieval effectiveness on a document collection. Each method consists of a combination of term scoring scheme, term scoring formula, and term assignment scheme. Our preliminary experiments evaluate these different slot-based query expansion methods on a collection of news documents,and conclude that:(1) slot-based queries yield better retrieval accuracy compared to keyword-based queries in the complex question problems; and (2)directly applying traditional query expansion on the query terms of each slot does not always work well.

Discipline

Databases and Information Systems | Numerical Analysis and Scientific Computing

Publication

Workshop on CyberInfrastructure: Information Management in eScience (CIMS2007), In conjunction with ACM 9th International Workshop on Web Information and Data Management (WIDM 2007)

Identifier

10.1145/1317353.1317364

Publisher

ACM

City or Country

Lisboa, Portugal

Additional URL

http://doi.org/10.1145/1317353.1317364

Share

COinS