Vachkov G.L.Tzanev A.T.2024-07-102024-07-102024-07-102024-07-102018-01-012405-896310.1016/j.ifacol.2018.11.297SCOPUS_ID:85057023402https://rlib.uctm.edu/handle/123456789/510Large data sets are often collected from the real time operation of complex machines and technological systems. Then the collected information is further used for different purposes, such as performance evaluation, anomaly detection and fault diagnosis of the complex systems. The first step of analyzing the large data sets is to discover the degree of similarity between preliminary given pairs of data sets. In this paper the similarity analysis of large data sets is performed by using the so called grid fuzzy models (GFM), combined with a procedure of fuzzy decision making. The GFM are defined over an incomplete grid in the multidimensional input space and the number of the fuzzy rules depends on the concrete distribution of the data sets in this input space. The GFM is used to generate two important characteristics of the data set that are further processed by the fuzzy decision unit in order to produce the similarity degree. The whole technology is illustrated on the example of real data stream from a continuous technological process.enConference Paper