Which Way for a C ass~ficatmn Scheme for Computers and Medicine Roy Rada The National Library of Medicine supports an extensive collection of online indexed document citations for the biomedical field, The indexing scheme, although one of the first of its kind, remains one of the most sophisticated in everyday usage. The 14,000 terms of the Medical Subject Headings (MESH) are organized in a hierarchy of "is-a" relations that constitute a very simplified knowledge-base of the biomedical literature. The MeSH classification of articles seems to provide better support of retrieval than the also popular string-searching strategies of some document retrieval systems (see D Blair and M Maron "An Evaluation of Retrieval Effectiveness for a Full-Text Document-Retrieval System" in Comm ACM, March 1985, p 289-299). In the intersection of computers and medicine MeSH is understandably sketchy. But as the progress in medicine becomes more connected to high-technology the need for a richer MeSH classification of computer-related medical work increases. A committee of scholars could convene, study MESH, and propose amendments that would refine its coverage of computer-type material. Alternatively, one could explore algorithms for merging the better parts of MeSH's coverage of computing with the better parts of a
/lp/association-for-computing-machinery/which-way-for-a-classification-scheme-for-computers-and-medicine-3fGDfhUOEr