Abstract
A study was performed to assess system and user attributes associated with successful searching. A group of highly experienced retrieval system users were given eight queries - half to be completed with a Boolean system and the other half to be completed with a natural language system - and asked to identify `instances' of information related to a topic. A variety of demographic, experience, cognitive, personality, search mechanics, and user satisfaction attributes were measured and evaluated for association with `instance recall.' The system-oriented analysis showed user preference for the Boolean system but equivalent instance recall with the two systems. The user-oriented analysis showed improved performance with occupation type, user satisfaction, and number of search terms per cycle. A variety of cognitive and personality attributes did not show any association with better searching.
Original language | English (US) |
---|---|
Pages | 60-70 |
Number of pages | 11 |
State | Published - 1999 |
Event | Proceedings of the 1999 IEEE Forum on Research and Technology Advances in Digital Libraries, IEE ADL '99 - Baltimore, MD, USA Duration: May 19 1999 → May 21 1999 |
Other
Other | Proceedings of the 1999 IEEE Forum on Research and Technology Advances in Digital Libraries, IEE ADL '99 |
---|---|
City | Baltimore, MD, USA |
Period | 5/19/99 → 5/21/99 |
ASJC Scopus subject areas
- Management of Technology and Innovation