Effectiveness of web page classification on finding list answers

List question answering (QA) offers a unique challenge in effectively and efficiently locating a complete set of distinct answers from huge corpora or the Web. In TREC-12, the median average F1 performance of list QA systems was only 6.9%. This paper exploits the wealth of freely available text and link structures on the Web to seek complete answers to list questions. We employ natural language parsing, web page classification and clustering to find reliable list answers. We also study the effectiveness of web page classification on both the recall and uniqueness of answers for web-based list QA. Categories and Subject Descriptors H.3.3 [Information Storage and Retrieval]: Information Search and Retrieval General Terms Algorithms, Performance Keywords Question Answering, Web page classification

• Type: Conference
• Year: 2004
• Where: SIGIR
• Authors: Hui Yang, Tat-Seng Chua
• Keywords: List Qa, Question Answering, SIGIR 2004, Web Page Classification