Data mining, Bongard problems, and the concept of pattern conception

Abstract

The concept of pattern conception was discussed for categorization of data mining architectures. The difference between pattern conception and pattern recognition models lead to the categorization of information processing architectures. The raw data was perceived in terms of specific 'higher-level' structures. The newly activated concepts created top-down pressure to search new possibilities. In long term memory, the activation of a concept propagated activation to related concepts.