Abstract
The article examines the problem of the spread of malicious advertising programs through web pages that pose a serious threat to the privacy and security of Internet users. Using machine learning algorithms to detect and neutralize malicious advertising programs embedded in Web pages. By focusing on data processing, tag extraction, and classification techniques, machine learning analyzes in detail how it can improve malware detection processes. Various machine learning algorithms, including logistic regression, decision trees, random forest, naive Bayesian and ensemble methods, are being studied to determine their effectiveness in distinguishing malicious and legitimate advertising content.A methodology for building training and test models, including data on malicious and secure advertising modules, is described. Various approaches to machine learning, including teacher-led learning, unsupervised learning, and deep learning techniques, are being analyzed to identify hidden patterns of harmful behavior. The results of the study show that the use of machine learning algorithms makes it possible to detect malicious advertising programs with high accuracy, which can become the basis for the development of more effective cybersecurity tools. Potential problems and limitations of existing methods are also discussed, as well as directions for further research on detecting malicious advertising programs using machine learning.