Knowledge-internalization process for neural-networks practitioners

Abstract

This study explores the knowledge-internalization process within which a neural-network practitioner embody the explicit knowledge obtained from extracting network`s preimage, the set of input values for a given output value, into his/her tacit knowledge. With a number of well-trained single-hidden layer feed-forward neural networks, the practitioner first extracts the (nonlinear) preimage of each trained network. The practitioner then internalizes the explicit outcomes and the insights obtained from the preimage extracting process into his/her tacit knowledge bases. We use the experiment of bond-pricing analysis to illustrate the knowledge-internalization process. This study adds to the literature by introducing the knowledge-internalization process. Moreover, in contrast to the data analyses in previous studies, this study uses mathematical analyses to identify networks` preimages.