DATA MINING IMPLEMENTATION FOR LIBRARIES

Abstract

The World Wide Web (WWW) is gaining popularity by providing an alternative way for individuals to get the information they require easily and quickly through the Internet. This is threatening the existence of libraries, which have been one of the main resources for information. The librarians in return are proactively planning and improving their services to meet the challenges imposed by the World Wide Web. Libraries are using the technologies of the Internet to provide new and better services to their patrons. Although the librarians are being proactive, they require information to make strategic decisions that would improve their services for their patrons. One way of getting this information is to perform data mining on the data captured in the libraries’ systems. Data mining or also known as knowledge discovery from databases is a process of getting useful information or knowledge by analysing the data for patterns. The capability of data mining to solve some of the organisation’s persistent and pressing issues has encouraged new information technology users to implement this system. The Extraction, Transformation and Loading (ETL) processes are required to extract, transform and load the data into a database for the data mining activities. In this dissertation, firstly, a survey was conducted on the use of data mining in the libraries of Higher Education Institutes and Universities in Malaysia. Secondly, a data extraction and data transformation application was developed for extraction, cleaning and transforming data for data mining. Finally, a prototype data mining system was implemented for the library integrating the developed data extraction and data transformation application with other suitable tools and methodology.