Contribution to the problem of knowledge discovery of a processional character in databases

The term KDD (knowledge discovery in databases) appeared in early 90s of the last century, due to the spread of information technology in the society and as a result of information technology development. The organization of large-scale, automated collection of data in databases, in data warehouses and data repositories enables non-trivial acquisition of implicit and potentially useful information from data. This issue is in practice mostly linked to the needs of strategic management decisions. In the present paper one of the possible approaches to knowledge discovery in long-term measurement of data of a processional character is pointed out. Authors of the paper, using a special method, evaluated the long-term measurements of the air temperature in two cities (Košice, Prague). The temperature data were scanned online and transmitted to the database via the internet. The resulting two sequences of temperatures are appropriately reorganized into the structure of a vector space with the inner product. By a suitable transformation of data vectors from the time domain to the frequency domain and by subsequent visualization it was shown that the environment of man, in terms of thermodynamics, is a non-linear dynamic system with the characteristics of chaos. Although the knowledge gained only confirmed in an exact way a priori expectations, the method described can be used universally across the class of processes with the observed dynamics.