COMPARISON OF THE EFFECTIVENESS OF MACHINE LEARNING ALGORITHMS IN DETECTING SPAM MESSAGES IN KAZAKH

Abstract

The article presents a comparative analysis of the performance of various machine learning algorithms for spam detection, with a special focus on their application to the Kazakh language. The methods under consideration include Bayesian spam filtering, k-nearest neighbors, support vector machines, and decision trees. Traditionally, spam has been used to promote products and services to potential customers. However, it has become a tool for hacking and spreading viruses. To solve this problem, scientists and researchers have proposed various methods for detecting and filtering spam. The following are the different categories of spam filtering methods: Case-based spam filtering methods; Content-based filtering methods; List-based filtering methods; Heuristic or rule-based spam filtering methods; Adaptive Spam filtering methods. The development and evaluation of various machine learning approaches for detecting spam in the Kazakh language may become a potential research problem. Our goal in solving these research problems is to enrich the current literature by offering machine learning models that can detect spam messages in Kazakh. Kazakh language, machine learning algorithms, efficiency.