УДОСКОНАЛЕННЯ МЕТОДУ ВИЯВЛЕННЯ ТА КЛАСТЕРІЗАЦІЇ ДЖЕРЕЛ НЕПРАВДИВОЇ ІНФОРМАЦІЇ.

The main purpose of evolutionary optimization is to find a combination of parameters (independent variables) that would help maximize or minimize the qualitative, quantitative, and probabilistic characteristics of the problem. Recently, integrated optimization methods have become very common, borrowing the basic principles of their work from wildlife. Researchers are experimenting with different types of representations, for example, evolutionary and genetic algorithms use selection methods and genetic operators. A large number of algorithms based on the swarm method are known. The artificial bee colony is an optimization method that mimics the behavior of bees, a specific application of cluster intelligence, the main feature of which is that it does not need to understand specific information about the problem, you just need to optimize the problem. Comparing inferiority with the help of the local optimization behavior of each person with an artificial bee finally leads to the appearance in the group of a global optimal value with a higher rate of convergence. The paper considers the method of solving the optimization problem based on modeling the behavior of the bee colony. Description of the model of the behavior of intelligence agents and forage agents, search mechanisms, and selection of positions in a given neighborhood. The general structure of the optimization process is given. Graphical results are also presented, which prove the possibility of the bee colony method to optimize the results, i.e. from all multiple sources of information, the bee colony method by optimization can significantly limit the number of information sources, identify a narrow range of sources that may be false information. Which in the future will allow you to more accurately identify sources with false information and block them. [ABSTRACT FROM AUTHOR]