ANALYSING USER BEHAVIOR DATA FOR MOBILE APP OPTIMIZATION

Abstract

In today’s competitive mobile app landscape, understanding user behavior is essential for improving performance, engagement, and retention. This research focuses on applying machine learning techniques—Random Forest (RF), Decision Trees (DT), and Support Vector Machines (SVM)—to analyze user interaction patterns within mobile applications. Data from user sessions, in-app activities, and engagement metrics are processed to identify factors driving user satisfaction and churn. Random Forest is used for feature selection and ranking, highlighting the most influential variables. Decision Trees offer interpretable insights into user navigation and behavior paths, while SVM is employed to classify user preferences and predict future actions. The insights gained help developers personalize app experiences, refine UI/UX design, and implement data-driven improvements to enhance user retention.