Discovering Frequent Access Patterns in a Digital Library Using Association Mining

Data Mining, also known as knowledge discovery in databases, has been recognized as a promising new area for database research. Mining frequent item sets in transactional databases, binary transaction tables, time series databases and many other kinds of databases have been an active research topic over the past few years. Frequent access pattern is a special case of sequential pattern in an application database which helps to make effective decisions in the respective problem domain.

Given a large database of book transactions in the library, where each transaction consists of book-id, name of the book, author, and other related fields, the problem is to mine the frequent access patterns of the user from the library databases. The outcome of the findings will help the management to take effective steps that will cater the needs of the user.

Apriori and FP-growth algorithms can mine the complete sets of frequent item sets. These two algorithms were implemented and the performance of the algorithms was studied. The result shows that FP-growth algorithm performs well compared to Apriori.