Advertising, promoting, and selling goods and services are common place in society. Social media, search engines, websites, and other platforms are all part of the current trend of digital marketing. Advertising companies have shifted their focus to internet platforms that make it easier to attract customers. As a result, demand for digital technology platforms is rising. Many kinds of advertisements use a variety of platforms with a strategy and goal in mind. In this work, click ads related to mobile devices are taken into consideration. The user is taken to a website after clicking on the click advertisement. Following the pay-per-click (PPC) system, the advertising networks of the advertisement pay the publisher based on the number of clicks that result in advertisers. But this payout adheres to the Click Fraud security flaw technique. Click fraud is the illegal activity of clicking on pay-per-click ads in order to increase publisher revenue or deplete advertising budgets. The use of man-made intelligence techniques to tackle intricate issues in a variety of academic domains, such as cyber security, has increased and can yield unexpected outcomes. To ascertain whether or not the user is fraudulent they use Random Forest, Decision Trees, and Logistic Regression. It is suggested in this work to identify frauds concealed in user clicks. To provide the optimal model, evaluation measures including accuracy, precision, and recall are taken into account out of the three algorithms. The use of logistic regression produced findings that were appropriate, thus the Logistic Regression model is used to classify ad click frauds.