The discovery of frequent patterns with logic and constraint programming

The basic goal of data mining is to discover patterns occurring in the databases, such as associations, classification models, sequential patterns, and so on. In this paper we focus on the problem of frequent pattern discovery, which is the process of searching for patterns such as sets of features or items that appear in data frequently. Such frequent patterns can reveal associations, correlations, and many other interesting relationships hidden in a database. Most of frequent pattern mining systems in the market are too generic and become inefficient when set of patterns is large and the frequent patterns are very long. A new trend in data mining is a scalable method that uses constraints to guide the system in its search for interesting patterns. Our main research objective is the development of constraint-based mining methodology and this paper presents the preliminary results of our study and prototype development. We present the implementation of frequent pattern mining system based on declarative programming paradigm using logic programming and constraint logic programming. The comparative performance studies on speed and memory usage of logic versus constraint programming are also reported in the paper.