Inferring user intent in web search by exploiting social annotations

In this paper, we present a folksonomy-based approach for implicit user intent extraction during a Web search process. We present a number of result re-ranking techniques based on this representation that can be applied to any Web search engine. We perform a user experiment the results of which indicate that this type of representation is better at context extraction than using the actual textual content of the document. Categories and Subject Descriptors H.3.3 Information Search and Retrieval – retrieval models, information filtering. General Terms Algorithms, Experimentation. Keywords Web search, folksonomy, context.

• Type: Conference
• Year: 2010
• Where: SIGIR
• Authors: Jose M. Conde, David Vallet, Pablo Castells
• Keywords: Information Management, SIGIR 2010, User Intent Extraction, Web Search, Web Search Process