Machine Learning-Based Classification of Online Users Using Information Seeking Behavior Analysis

Abstract

This paper presents a Machine Learning approach for the classification of online users by analyzing and exploiting their information-seeking behavior on digital platforms. With the rapid growth of internet usage, understanding user behavior has become essential for improving personalized services, targeted recommendations, and online security. Information-seeking behavior includes patterns such as search queries, browsing history, clickstream data, time spent on pages, and navigation sequences. By collecting and processing these behavioral attributes, machine learning algorithms can effectively classify users into different categories based on their interests, intentions, and interaction styles. The proposed system applies data preprocessing, feature extraction, and classification techniques using algorithms such as Decision Trees, Support Vector Machines, Random Forest, and Neural Networks to improve prediction accuracy. Experimental results demonstrate that behavioral patterns provide significant insights for accurate user classification and enhance system performance compared to traditional demographic-based methods. This approach can be widely applied in e-commerce, education platforms, social media analysis, and cybersecurity systems to deliver intelligent and adaptive user experiences.