An expert selector of online public access catalogues.

Blackadder, Alistair

Abstract

This project looks at the possibility of creating an expert system which would aid with access to OPACs available on the JANET network. This system would aid access by being able to select the OPAC most suited for a search in a particular subject area. Looked at initially are the OPACs available and the JANET network. The problems with access and use of these OPACs are demonstrated. Expert systems and how they operate are looked at in relation to how an expert system might be used to solve these problems. The uses already made of expert systems within library and information services are examined in order to establish the potential of such a system. Examples are given of systems in order to demonstrate this potential. The differences between an expert selector and an inexpert selector are then examined. The system objectives required to make such a system "expert” are defined. Tools for constructing an expert system are detailed. The different options are considered and the selection of the expert system shell Crystal is examined. The knowledge required to construct an expert system is then examined. Problems in knowledge acquisition are considered and the knowledge to be used in this project is evaluated, particularly the Scottish Conspectus project. Details are given of how this knowledge is used to create an expert system using the Crystal expert system shell. How the system operates and how it would communicate with the OPACs available through JANET is examined. The conclusion is reached that an expanded version of the demonstration system produced as a part of this project could be a valuable tool in aiding access to sources of information that are currently underused.

Citation

BLACKADDER, A. 1992. An expert selector of online public access catalogues. Robert Gordon University, MPhil thesis. Hosted on OpenAIR [online]. Available from: https://doi.org/10.48526/rgu-wt-2807301

• Thesis Type: Thesis
• Deposit Date: May 23, 2025
• Publicly Available Date: May 23, 2025
• DOI: https://doi.org/10.48526/rgu-wt-2807301
• Public URL: https://rgu-repository.worktribe.com/output/2807301
• Award Date: May 31, 1992