Increasing Importance of Analog Data Processing

Abstract

This paper points out that although DX (Digital Transformation) is getting wide attention these days and it is very good from the standpoint of making the most of the current bit computer, we should remember that most data the Natural World is composed of are analog. And analog data are increasing their importance quickly, because materials are getting softer and softer with the remarkable progress of material engineering. We need need to directly interact with the objects to find out what they are and how we should handle them. And although our traditional engineering approach was Euclidean Space based, with orthonormality and interval-based distance with units as its requirements, the Real World is getting more and more complicated and complex, and the data come to contain a wide variety of information. So, the Euclidean space approach cannot be applied anymore and we need to develop Non-Euclidean Space approach. To cope with this situation, Mahalanobis Distance, which is an ordinal approach combined with pattern is proposed. The basic idea of this approach is to help our Instinct with performance indicator to provide it with the whole picture of the situation and to enable it to make the appropriate decisions on what actions we should take. As this approach is Instinct-based, it can process data with extremely large number of dimensions.