A Parallel Architecture for Data Mining and Knowledge Discovery

In this paper, we introduce a new machine learning theory based on multi-channel parallel adaptation for rule discovery. This theory is distinguished from the familiar parallel-distributed adaptation theory of neural networks in terms of channel-based convergence to the target rules. We show how to realize this theory in a learning system named CFRule. CFRule is a parallel weight-based model, but it departs from traditional neural computing in that its internal knowledge is comprehensible. Furthermore, when the model converges upon training, each channel converges to a target rule. The model adaptation rule is derived by multi-level parallel weight optimization based on gradient descent. Since, however, gradient descent only guarantees local optimization, a multi-channel regression-based optimization strategy is developed to effectively deal with this problem. Formally, we prove that the CFRule model can explicitly and precisely encode any given rule set. Also, we prove a property related to asynchronous parallel convergence, which is a critical element of the multi-channel parallel adaptation theory for rule learning. Thanks to the quantizability nature of the CFRule model, rules can be extracted completely and soundly via a threshold-based mechanism. Finally, the practical application of the theory is demonstrated in DNA promoter recognition and hepatitis prognosis prediction.