Experimental Study of Kernel Density Functions in Angelov–Yager Fuzzy Classification and Clustering Systems during the Processing of Streaming Data to Solve Cybersecurity Problems
摘要
The efficient elimination of cyberthreats implies the detection of attacks to take real-time protective measures on the basis of streaming data, but the existing solutions do not always capable of data processing in such a mode. To solve these problems, it is proposed to apply the incremental learning method, i.e., Angelov–Yager first-order fuzzy classification and clustering systems. In the original implementation, the Cauchy kernel function is applied, and the use of alternative kernel functions has not been considered before. Alternative kernel functions on the basis of cybersecurity datasets were experimentally studied, the statistically significant difference in the accuracy and number of rules in the application of different functions were revealed, and the best kernel functions for individual datasets were determined