Shu‐Chuan Liao, Kuo‐Fong Kao, I‐En Liao, Hui‐Lin Chen and Shu‐O Huang
As library collections increase rapidly, personalized recommender systems have become a very important service for library patrons. The purpose of this paper is to design and…
Abstract
Purpose
As library collections increase rapidly, personalized recommender systems have become a very important service for library patrons. The purpose of this paper is to design and implement a personal ontology recommender (PORE) system by building personal ontologies based on patrons' borrowing records.
Design/methodology/approach
In the PORE system, the traditional cataloging scheme, classification for Chinese libraries, is used as the reference ontology. This reference ontology is transformed to a unique personal ontology for each user based on the mining results from library borrowing records of that user.
Findings
A personal ontology represents a unique user interest on specific subjects. The personal ontology can be used to filter out unsuitable recommendations based only on a keyword matching method. Besides, the recommended books can be organized into the personal ontology, and provide the patron with a user‐friendly interface to access library collections.
Research limitations/implications
The PORE system is currently implemented for Chinese collections. From this paper builds a new version to support English collections by adopting the Library of Congress Classification as the reference ontology.
Originality/value
This paper represents a practical method of building a user's personal ontology and explains the functional use of ontology knowledge.