Application of the Mechanisms of the Artificial Immune System to Solve the Problem of Detecting Anomalies in the Data Series

Abstract

Biology has always been a source of ideas for various sciences. Computer science is no exception. For example, neural networks and genetic algorithms have already become widespread. In recent decades, I have been actively paying attention to the study of the mechanisms of the immune system. Its principles of operation have wide application in a number of optimization tasks, data processing and analysis, and many others. Detecting anomalies in the behavior of a system or processes is an important task for many applications, the task is to study the features of using this mechanism, implement it programmatically and analyze its operation.
The analysis of these works showed that these methods show good results, but they have a number of disadvantages, the most significant of which is the complexity of the organization and the long time spent on training. Thus, the problem is posed of developing new algorithms that are comparable in decision-making speed with artificial networks and expert systems and at the same time have less learning time. One of the ways to solve this problem is to develop a decision-making model based on an artificial immune system.The aim of the work is to develop and study a model for formalizing the decision-making process in the search for data anomalies using an artificial immune system.
The normal behavior of the system is often characterized by discrete time series of observations. In this case, the problem of finding anomalies can be formulated as the problem of finding unacceptable deviations in the characteristics of the system.
The artificial immune system represents an idealized version of the natural analogue and reproduces the key components of the natural process: selection of the best antibodies of the population depending on the degree of their affinity (proximity) to the antigen, cloning of antibodies, mutation of antibodies.