Autolearner: An Autonomic Machine Learning System Based on Concept Algebra

On the basis of the basic research in cognitive informatics, particularly the development of concept algebra, real-time process algebra (RTPA), the layered reference model of the brain (LRMB), and the object-attribute-relation (OAR) model for internal knowledge representation, the revilement of the cognitive process of learning and formal knowledge manipulation are enabled. This paper presents an autonomic learning system known as the AutoLearner. Mimics of knowledge organization, updating, and navigation inside the brain are formally modeled according to the OAR model using concept algebra and RTPA. A machine learning system and a cognitive simulator are developed to visualize interactions between thinking, learning, and the internal knowledge representation. A case study is presented to demonstrate the design and implementation of the AutoLearner system.