Information reuse and integration in artificial neural networks

The need to reuse information is urgent, and a shift is required in the development (understanding - research) of methodologies, with a more reuse-centric view leading to more effective knowledge integration, within a framework of knowledge actualisation and management. This paper describes a connectionist architecture (framework) and its rationale, in which knowledge embedded in one network may be reused in another. This allows information reuse and integration (inheritance) in the context of information acquired by a neural net. The paper focuses on early (initial) results; some of the aims have been demonstrated and amplified through the experimental work. This also enables us to assess the strength and weakness of the approach. It concludes that the underpinning concepts - inheritance and transformation - are viable and demonstrate the basic feasibility of the architecture (and framework).