Jacques Ducloy

WebStress: A web interface to explore a multidatabase bibliographic corpus on occupational stress

A bibliographic study on a multidisciplinary subject, such as occupational stress, requires a multidatabase query facility. Given this, we searched eight databases belonging to different fields (medicine, psychology, etc.) and from different countries. Thus, more than 26000 references on occupational stress were gathered. A bibliographic corpus of this sort is unusable because of the heterogeneous format of the records and the number of duplicate entries. Considering the limitations of commercial programs, we have developed a system, called WebStress, for reformatting the data and removing duplicates. The huge number of references and the research needs of updating and searching have necessitated the addition of several other functions to WebStress. Therefore WebStress comprises a web interface that makes it possible to search the corpus with advanced features (using clustering, a specialized thesaurus on stress and Boolean queries). In addition to the usual bibliographic queries on a specific database such as Medline, WebStress provides bibliometric analysis of the corpus, which might contribute to a detailed analysis on occupational stress in order to highlight the networks of researchers and to find the main topics studied in this area. WebStress is not yet available to researchers in general. It is being further developed, and also permission needs to be obtained from the eight commercial bibliographic databases that it consults.

A Workbench for Bibliographic or Factual Data Handling

DILIB (Document and Information LIBrary) is a workbench for Scientific or Technical Information and Document Engineering. It has been designed in order to make investigations on heterogeneous sets of data or to build Information Retrieval Applications. Its kernel is a toolkit whose basic part consists in a SGML tree handling library and which contains basic elements for building Information Retrieval Systems. We present two aspects of its utilization: building WWW hypertexts from bibliographic databases, and developing an experimental IRS system, NOMAD, based on a symbolico-connectionist approach.