The minimum description length principle for pattern mining: a survey

Abstract Mining patterns is a core task in data analysis and, beyond issues of efficient enumeration, the selection constitutes major challenge. The Minimum Description Length (MDL) principle, model method grounded information theory, has been applied to pattern mining with aim obtain compact high-quality sets patterns. After giving an outline relevant concepts from theory and coding, we review MDL-based methods for different kinds various types data. Finally, open discussion on some regarding these methods.