Improving user performance on Boolean queries

ABSTRACT

The accurate formulation of boolean expressions is a notorious problem in programming languages as well as database and web query tools. Users have demonstrated great difficulty with the common textual method for specifying these queries, which uses the boolean operators AND, OR, and NOT, partly because these words are used inconsistently in natural languages. This paper proposes a tabular boolean query language that avoids the need to use named operators, provides a concrete distinction between conjunction and disjunction, and makes grouping more explicit. A study comparing this tabular language with textual boolean expressions found that untrained users perform better when they express their queries in the tabular language, and about equally well when interpreting queries written in either language. We conclude that systems can benefit by adopting a tabular notation for query formulation.

REFERENCES

• 1: P. G. Anick , J. D. Brennan , R. A. Flynn , D. R. Hanssen , B. Alvey , J. M. Robbins, A direct manipulation interface for boolean information retrieval via natural language query, Proceedings of the 13th annual international ACM SIGIR conference on Research and development in information retrieval, p.135-150, September 05-07, 1990, Brussels, Belgium [doi>10.1145/96749.98015]
• 2: S. L. Greene , S. J. Devlin , P. E. Cannata , L. M. Gomez, No IFs, ANDs, or ORs: a study of databases querying, International Journal of Man-Machine Studies, v.32 n.3, p.303-326, March 1990 [doi>10.1016/S0020-7373(08)80005-3]
• 3: McQuire, A. and Eastman, C. M. Ambiguity of Negation in Natural Language Queries. In Proceedings of the Eighteenth Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. Posters: Abstracts. (1995), 373.
• 4: Pane, J. E and Myers, B. A. Natural and Accurate Ways to Specify the Selection of Objects from a Group. submitted for publication (2000), http://www.cs.cmu.edu/~pane/study3.html.
• 5: Tanaka, J., "The Perfect Search" in Newsweek, 134, 13, (September 27 1999), 71.

INDEX TERMS

Keywords:
end-user programming, information retrieval, psychology of programming, user studies