From research to application: the cite natural language information retrieval system

  • Authors:
  • Tamas E. Doszkocs

  • Affiliations:
  • National Library of Medicine

  • Venue:
  • SIGIR '82 Proceedings of the 5th annual ACM conference on Research and development in information retrieval
  • Year:
  • 1982

Quantified Score

Hi-index 0.00

Visualization

Abstract

Large operational information retrieval systems typically employ inverted file structures and Boolean logic operators for efficient text retrieval. These systems require considerable user training for effective use. As a consequence, searching is commonly performed by professional intermediaries on behalf of end users.By contrast, many small scale experimental retrieval systems incorporate desirable user interface features, such as natural (English) language querying, ranked output and relevance feedback.The author describes the design and implementation of a natural language search interface to MEDLINE, the National library of Medicines largest and most heavily used data base. The CITE (Current Information Transfer in English) prototype system is a large-scale, weighted logic information retrieval system with natural language query input, ranked search output, dynamic user feedback and automatic associative vocabulary mapping capahilities.