Self-organizing systems for knowledge discovery in large databases

We present a framework in which self-organizing systems can be used to perform change of representation on knowledge discovery problems, to learn from very large databases. Clustering using self-organizing maps is applied to produce multiple, intermediate training targets that are used to define a new supervised learning and mixture estimation problem. The input data is partitioned using a state space search over subdivisions of attributes, to which self-organizing maps are applied to the input data as restricted to a subset of input attributes. This approach yields the variance-reducing benefits of techniques such as stacked generalization, but uses self-organizing systems to discover factorial (modular) structure among abstract learning targets. This research demonstrates the feasibility of applying such structure in very large databases to build a mixture of ANNs for data mining and KDD. Areas of applications include multi-attribute risk assessment using insurance policy data, text document categorization, and anomaly detection.