Modeling cataloging expertise: A feasibility study

摘要

The development of expert systems for cataloging has had limited success to date. Although there has been some promise in the area of advisory or training systems, no system has been able to achieve satisfactory performance in practice. Much of the blame has been placed on two issues: the ambiguity and lack of compatible computerized structure of Anglo-American Cataloging Rules and the lack of understanding of expertise possessed by experienced catalogers. A project attempting to address the latter issue was planned in cooperation with National Agricultural Library. This paper describes the findings from the feasibility study. The project concentrates on the nature of cataloging expertise by attempting to answer the following research questions: (a) what tasks of cataloging involve expertise? (b) what are the components of cataloging expertise? and (c) what distinguishes the work performed by beginning catalogers from that performed by expert catalogers? The results of the study suggest that tasks involving cataloging expertise can be found in the following groups: (a) searching bibliographic databases, (b) determining access points, (c) interpreting bibliographic data and rules, (d) identifying and prioritizing problems and special cases, and (e) administering performance. The following facets of skill are further identified: (a) technology, such as online retrieval and knowledge about online catalogs; (b) professional tools, such as cataloging codes and related reference tools; (c) communication, such as knowledge about vendors and about operations in other departments; and (d) subject specialty and languages. A model is suggested, using task categories and skill facets as two dimensions, for representing the components of cataloging expertise. The plan of conduct for the cataloging expertise project is also described.