Design of a Study to Evaluate the Effectiveness of a Spoken Language Interface to Information Systems

摘要

Users of digital libraries and other information systemstypically carry out searches with very short queries, on theorder of two words or so. This makes it very difficult for thesystems to disambiguate their queries and identify potentiallyrelevant documents, resulting in sub-optimal retrievalperformance. We hypothesize that users will provide betterand more useful descriptions of their information problems ifthey are able to speak to the system and easily indicatethrough speech and gesture those documents and aspects ofdocuments which they find useful, and not useful. In thispaper, a spoken interface is described, and a planned wizardof oz study will be introduced.