An Empirical Comparison of Techniques for Selecting and Combining Local Patterns into a Global Model

Local pattern discovery, pattern set discovery and global modeling build together as consecutive steps a specific case of global pattern discovery. As each of these three steps have gained an increased attention in recent years, a great variety of techniques for each step have been proposed. Though so far there has been no systematic comparison of the possible choices. In this paper, we will consider a special representative of local patterns, namely class association rules, and evaluate several options for pattern set discovery and for global modeling for this type of classification rules