CCE Theses and Dissertations
Turning an Expert Reference Librarians' Skill Into computer Software: A Practical Application of Artificial Intelligence Techniques
Date of Award
1987
Document Type
Dissertation
Degree Name
Doctor of Arts (DA)
Department
Center for Computer-Based Learning
Advisor
Jane Anne Hannigan
Committee Member
Marlyn Kemper Littman
Committee Member
Berry Centini
Abstract
Literature and research in the areas of£ question negotiation and reference direction were gleaned for sets of generally accepted rules about a professional reference librarian’s skills. The sets of rules allowed the development of a rule-based system to address an important function in libraries - providing reference assistance to patrons.
An expert system is a program that takes the place of human experts by codifying the knowledge and rules they use to reach their conclusions. A computer equipped with an expert's knowledge has the ability, like a good assistant, to do things for the library in the way the reference librarian would do them. This looms as a potential revolution in the productivity of the library because it would free personnel to accomplish other tasks.
The goal of this project was to develop and implement a knowledge-based system to automate the reference process in libraries. The system was based upon library experts who worked with standard reference tools on typical reference questions encountered in libraries. The system was localized to be specific to the reference collection of St. Thomas Aquinas High School in Fort Lauderdale, FL. The expert system was evaluated as a success by producing advice that was equivalent to or an improvement upon that given by its human counterparts on a given set of questions.
The results of the evaluation indicated that the expert system would produce results that are not only more reliable, but also more efficient than those of a reference librarian. While the librarian often directed patrons to the first source of information that occurred to him/her, the computer, on the other hand, quickly evaluated and selected the best sources available. An expert system developed as described would not only reduce the time necessary to answer questions, but also increase the number of sources it cites as appropriate to the reference problem area.
Relevant research and a complete knowledge-based program are included.
NSUWorks Citation
Paul L. Catano. 1987. Turning an Expert Reference Librarians' Skill Into computer Software: A Practical Application of Artificial Intelligence Techniques. Doctoral dissertation. Nova Southeastern University. Retrieved from NSUWorks, Center for Computer-Based Learning. (447)
https://nsuworks.nova.edu/gscis_etd/447.