The literary works of Sir William Osler

A paper read at the first regular meeting of the Osler Society of McGill University, October 19, 1921.     

A description of the catalog division project at the College of Physicians of Philadelphia Library.

This paper describes the procedures used at the Library of the College of Physicians of Philadelphia to divide its ninety-year-old dictionary card catalog. The division was necessitated by overcrowding, obsolete subject headings, and lack of a complete authority list which resulted in like materials being scattered throughout the catalog under several headings. Two catalogs were created: the historical-biographical catalog, representing all works published before 1950 and all works of historical or biographical nature; and the current catalog, containing all works published from 1950 on, excepting historical or biographical materials. The 1950- catalog was further divided into name and subject catalogs, and the subject section was revised according to MeSH. The project was completed in about two years. As a result, searching time has been much reduced, and the library is able to take advantage of the annual revisions of MeSH to update the subject catalog.     

A computer-generated catalog of audiovisuals.

A computer-generated catalog of nonprint media is described. Examples are given of four access points to the data base: (1) main entry, (2) title, (3) MeSH terms, and (4) broad subject categories. The data input procedure is summarized. The AV catalog as the basis for a union list is evaluated, and finally, the catalog in relation to certain local problems is discussed.     

Construction of query concepts based on feature clustering of documents

In Information Retrieval, since it is hard to identify users’ information needs, many approaches have been tried to solve this problem by expanding initial queries and reweighting the terms in the expanded queries using users’ relevance judgments. Although relevance feedback is most effective when relevance information about retrieved documents is provided by users, it is not always available. Another solution is to use correlated terms for query expansion. The main problem with this approach is how to construct the term-term correlations that can be used effectively to improve retrieval performance. In this study, we try to construct query concepts that denote users’ information needs from a document space, rather than to reformulate initial queries using the term correlations and/or users’ relevance feedback. To form query concepts, we extract features from each document, and then cluster the features into primitive concepts that are then used to form query concepts. Experiments are performed on the Associated Press (AP) dataset taken from the TREC collection. The experimental evaluation shows that our proposed framework called QCM (Query Concept Method) outperforms baseline probabilistic retrieval model on TREC retrieval.