In the modern world the basis of the development is to obtain the information, and the appropriate information processing. Thanks to the technical developments, more and more part of the life can be observed to collect data. However, massive accumulation of data not so useful, it is necessary to extract the relevant information. This is the task of the statistics.
The purpose of my thesis is to present different kind of data mining and statistical methods at circuit level. At the design part it was important to minimize the resources and handle large amounts of data. During the development I analyzed necessary data quantity for the algorithms and feasibility of the operations and where was necessary, the structure of the algorithm was modified.
My task was to plan instruction execution unit for processing special kind of tasks. The unit is able to perform different kind of hierarchical clustering methods and the DBSCAN algorithm and able to calculate Cramer's association coefficient, Pearson's contingency coefficient and Csuprov's T-index. In my thesis I also worked on the implementation of principal component analysis, but it was not successful because of the data representation method.