What Makes a Good Answer? The Role of Context in Question Answering

Question answering systems have proven to be helpful to users because they can provide succinct answers that do not require users to wade through a large number of documents. However, despite recent advances in the underlying question answering technology, the problem of designing effective interfaces has been largely unexplored. We conducted a user study to investigate this area and discovered that, overall, users prefer paragraph-sized chunks of text over just an exact phrase as the answer to their questions. Furthermore, users generally prefer answers embedded in context, regardless of the perceived reliability of the source documents. When researching a topic, increasing the amount of text returned to users significantly decreases the number of queries that they pose to the system, suggesting that users utilize supporting text to answer related questions. We believe that these results can serve to guide future developments in question answering interfaces.

[1]  Ellen M. Voorhees,et al.  Overview of the TREC 2002 Question Answering Track , 2003, TREC.

[2]  Eduard H. Hovy,et al.  Using Knowledge to Facilitate Factoid Answer Pinpointing , 2002, COLING.

[3]  Jimmy J. Lin,et al.  The START Multimedia Information System: Current Technology and Future Directions , 2002, Multimedia Information Systems.

[4]  Jimmy J. Lin,et al.  Data-Intensive Question Answering , 2001, TREC.

[5]  Ryen W. White,et al.  Finding relevant documents using top ranking sentences: an evaluation of two alternative schemes , 2002, SIGIR '02.

[6]  Ellen M. Voorhees,et al.  The eleventh text REtrieval conference, TREC 2002 , 2003 .

[7]  Andrew Turpin,et al.  Do batch and user evaluations give the same results? , 2000, SIGIR '00.

[8]  Jinwoo Kim,et al.  Effects of contextual navigation aids on browsing diverse Web systems , 2000, CHI.

[9]  Sanda M. Harabagiu,et al.  Performance Issues and Error Analysis in an Open-Domain Question Answering System , 2002, ACL.

[10]  Boris Katz,et al.  Annotating the World Wide Web using Natural Language , 1997, RIAO.

[11]  Gary Marchionini,et al.  Previews and overviews in digital libraries: designing surrogates to support visual information seeking , 2000 .

[12]  Mark D. Apperley,et al.  A review and taxonomy of distortion-oriented presentation techniques , 1994, TCHI.

[13]  Susan T. Dumais,et al.  Optimizing search by showing results in context , 2001, CHI.

[14]  Nicholas J. Belkin,et al.  Support for Question-Answering in Interactive Information Retrieval: Rutgers' TREC-9 Interactive Track Experience , 2000, TREC.