Abstract
In this paper we propose and evaluate interfaces for presenting the results of web searches. Sentences, taken from the top retrieved documents, are used as fine-grained representations of document content and, when combined in a ranked list, to provide a query-specific overview of the set of retrieved documents. Current search engine interfaces assume users examine such results document-by-document. In contrast our approach groups, ranks and presents the contents of the top ranked document set. We evaluate our hypotheses that the use of such an approach can lead to more effective web searching and to increased user satisfaction. Our evaluation, with real users and different types of information seeking scenario, showed, with statistical significance, that these hypotheses hold.
Original language | English |
---|---|
Pages | 213-220 |
Number of pages | 7 |
Publication status | Published - 1 Sept 2003 |
Event | 9th IFIP TC13 International Conference on Human-Computer Interaction. Interact 2003 - Zurich, Switzerland Duration: 1 Sept 2003 → 5 Sept 2003 |
Conference
Conference | 9th IFIP TC13 International Conference on Human-Computer Interaction. Interact 2003 |
---|---|
Country/Territory | Switzerland |
City | Zurich |
Period | 1/09/03 → 5/09/03 |
Keywords
- web search
- interfaces
- user studies
- information retrieval