Sage Journals: Discover world-class research

Abstract

A method of using relevance judgements to modify docu ment descriptions has previously been proposed. An experi ment is described which investigates the effect that application of this method, based on one query, has on a second query. The results indicate that approximately half of the second queries have substantially different retrieved document sets when docu ment modification is used. Of these, about half behave better, and half worse, than originally. If the queries are repeated a second time, the proportion showing improved performance increases to 90% and only 3% still give inferior retrieved document sets. The general conclusion is that document mod ification can lead to improved information system perfor mance.

Get full access to this article

View all access options for this article.

References

A. Bookstein and C. Kraft , Operations research applied to document indexing and retrieval decisions, J. ACM 24(3) (1977) 418-427.

C.H. Davis and J.E. Rush , Guide to Information Science (Greenwood Press, Connecticut, 1979) 91-97.

L.M. Purgailis Parker , Factors affecting document learning, Proc. ASIS Annual Meeting, 19 ( 1982).

L.M. Purgailis Parker , Towards a theory of document learning , J. Amer. Soc. Information Sci. 34(1) (1983) 16-21.

S.E. Robertson and K. Sparck Jones , Relevance weighting of search terms, J. Amer. Soc. Information Sci. 27 ( 1976) 129-146.

G. Salton and M.J. McGill , Introduction to Modern Information Retrieval (McGraw-Hill , New York, 1983) 145-146.

K. Sparck Jones and R.G. Bates , Research on Automatic Indexing 1974-1976 , Vol. 1: Text (University of Cambridge Press, Cambridge, 1977 ) Chap. 2.

An experiment in document representative modification

Abstract

Get full access to this article

References