General Frameworks for Combined Mining: Discovering Informative Knowledge in Complex Data

Enterprise data mining applications such as mining government service data often involve multiple large heterogeneous data sources, user preferences and business impact. Business people expect data mining deliverables to inform direct business decision-making actions. In such situations, a single method or one-step mining is often limited in discovering informative knowledge. It would also be very time and space consuming, if not impossible, to join relevant large data sources for mining patterns consisting of multiple aspects of information. It is crucial to develop effective approaches for mining patterns combining necessary information from multiple relevant business lines, catering for real business settings and delivering decision-making actions rather than providing a single line of patterns. The recent years have seen increasing efforts on mining such patterns, for example, integrating frequent pattern mining with classifications to generate frequent pattern-based classifiers. Rather than presenting a specific algorithm, this paper builds on our existing works and proposes combined mining as a general approach to mining for informative patterns combining components from either multiple datasets or multiple features, or by multiple methods on demand. We summarize general frameworks, paradigms and basic processes for multi-feature combined mining, multi-source combined mining and multi-method combined mining. Several novel types of combined patterns such as incremental cluster patterns result from such frameworks, which cannot be directly produced by existing methods. Several real-world case studies are briefed which identify combined patterns for informing government debt prevention and improving government service objectives. They show the flexibility and instantiation capability of combined mining in discovering more informative and actionable patterns in complex data. We also present combined patterns in dynamic charts, a novel pattern presentation method reflecting the evolution and impact change of a cluster of combined patterns and supporting business to take actions on the deliverables for intervention.