Application of Incremental Mining and Apriori Algorithm on Library Transactional Database

Abstract

Data mining is used to extract hidden, predictive information from large databases, which can be used for predicting future trends and allowing businesses to make knowledge-driven decisions [1].In this paper we explain how Apriori algorithm can be applied on the university's library transactional database in order to find out the frequent book items and generate rules on these book items so as to predict the book borrowing behavior of the students. It then explains how incremental mining when incorporated by adding five more transactions to the original set of ten transactions changes the number of frequent item-sets and association rules generated by the algorithm.