Proceedings of 14th International Conference on Conceptual Structures (ICCS 06), Aalborg, Denmark, 2006.
K.E. Wolff, Tobias Wolff
The purpose of this paper is to make explicit basic differences between the classical statistical approach and the conceptual approach as it is used in Conceptual Knowledge Processing. These two approaches differ deeply with respect to their strategies for understanding data. In contrast to the statistical approach of information reduction to a few numbers like mean and variance, the conceptual approach tries to unfold the data. The two approaches also differ in the usage of their mathematical tools, as for example the usage of numbers and concepts, of functions and relations, especially in the usage of order relations. They also differ by their formal treatment of “distributed objects”. These differences are made explicit using a case study in image processing on coding artifacts recently done by the second author.Read/download now