The Method of Developing a Classifier Using the Bayes Theorem for Making a Decision on the Determination of True Information

Abstract

The range of application of cluster analysis is very wide: it is used in archeology, medicine, psychology, biology, public administration, regional economy, marketing, sociology and other disciplines. Each discipline has its own requirements for primary data and rules for forming groups. Obviously, there will be different methodological approaches to market segmentation, the purpose of which is to identify groups of objects that are similar in terms of features and properties and to the formation of clusters that unite to strengthen their competitive advantages. Thus, when processing information in the information space, the methodology is usually aimed at building a mathematical model of cluster analysis of the object or phenomenon under study, and even obtaining an answer to the question: "Is the information true or not." Detecting false information in the digital world is an important task in overcoming the widespread spread of rumors and prejudices. The paper analyzes the existing methods of information classification in the information age. Formulate the signs of the information age, in the context of determining the veracity of information. Based on the main features of the information age, a method of creating a classifier has been developed to solve the problems of determining the veracity of information. Mathematical modeling was carried out using the developed classifier to confirm the developed method of decision-making about the veracity of information using the Bayes theorem. The obtained results proved the efficiency of the proposed method of developing a classifier for which, when applying the Bayes theorem for decision-making, it is possible to determine the veracity of information. But the developed Bayesian classifier is based on the fact that the a priori probabilities of the hypotheses are known. Therefore, the direction of further research is the development or improvement of methods and algorithms for determining the a priori probability of hypotheses.