Data mining tasks and methods: Rule discovery: characteristic rules

Descriptive data mining is the description of a set of data in a concise and summary manner and the presentation of the general properties of the data. Mining characteristic rules and discriminant rules from the data are two essential components in descriptive data mining. In contrast to online analytical processing, data description puts more emphasis on (1) automated processing, helping users determine which dimensions (or attributes) should be included in the analysis and to what abstraction level the data set should be generalized in order to obtain interesting summarization; and (2) handling complex data types. Mining data characteristics and discriminant descriptions can be implemented based on a data cube method or an attribute-oriented induction method. Moreover, data description can be enhanced by data dispersion analysis, multifeature data cubes, and discovery-driven data cubes.