In the present digital era, where cyber threats are increasing in both occurrence and complexity, the capability to analyze user behavior and identify risks associated with web browsing has become an essential aspect of modern cybersecurity (Vert et al., Introduction to Contextual Processing – Theory and Application. CRC Press, 2024). As the Internet becomes integral to communication, information retrieval, and business operations, individuals face growing risks from online threats such as malware, phishing attacks, and various forms of cybercrime (Iyengar et al., 2024). This study introduces an innovative approach to analyzing user behavior through browser history data, aiming to classify users into two categories: normal and abnormal. The classification framework is designed to provide actionable insights into online threats by examining patterns and activities recorded in users’ browsing history. The proposed approach leverages advanced machine learning techniques, specifically focusing on the Random Forest and XGBoost algorithms, to assess user-accessed URLs and determine their risk levels (Shi and Iyengar, Mathematical Theories of Machine Learning – Theory and Applications. Springer. ISBN: 978-30-30170-76-9, 2024). By assigning a risk score to each URL, this methodology enhances cybersecurity measures by enabling a more precise classification of user behavior. The system incorporates supervised learning techniques based on parameters such as protocol type, domain age, URL length, and suspicious keyword presence, increasing the accuracy of detecting and classifying potentially harmful URLs. The system has an intuitive interface for smooth interaction to guarantee usability and accessibility. A simple process allows users to upload their surfing data, which is then processed by the system to produce thorough reports automatically. These reports emphasize important findings, such as user classifications, risk assessments, and statistical analysis of browsing behaviors. These reports are enhanced with dynamic graphics made with the D3.js package to improve interpretability. Bar charts, pie charts, and other graphical components are examples of these visualizations, which offer an interactive and understandable depiction of URL statistics and related risk levels. The visualization dashboard, a crucial component of the project, enables users to interactively examine their surfing habits. The dashboard makes it easier to spot patterns that point to unusual or maybe harmful conduct by displaying surfing data in an organized and eye-catching way. Users can better understand their online behaviors with features like bubble charts that show domainwise risk distributions, heatmaps that show temporal patterns of user activity, and comparison bar charts for frequency analysis. The dashboard also offers tailored suggestions for reducing the hazards connected to risky browsing behaviors, such as staying away from high-risk websites or switching to safe surfing techniques. The importance of combining machine learning with visualization tools for proactive threat detection and user engagement is emphasized by this study. Through the integration of automated classification and real-time visualization, the solution helps to improve cybersecurity practices for both individuals and companies by providing a more thorough understanding of user behavior. Future developments in user behavior analysis, URL risk assessment, and interactive cybersecurity tools are made possible by the conclusions and contributions presented in this work, which will ultimately promote a safer digital environment in a world that is becoming more interconnected.