Enhanced Security Framework for Web Log Mining to Safeguard Organizational Data and Predict Consumer Behaviour

Abstract

Web log mining is a critical process for extracting meaningful patterns from web usage data, enabling organizations to enhance user experience and predict consumer behaviour. However, ensuring the security of web logs is paramount to protect sensitive information from malicious access and cyber threats. This paper proposes a comprehensive security algorithm designed to filter malicious web logs at the entry point, ensuring that only legitimate data is processed. The algorithm integrates three key phases: initial security check, data pre-processing, and session identification. After rigorous application of these algorithms, we achieved enhanced data integrity and identified patterns that contribute to accurate consumer behaviour predictions. This framework not only fortifies web log data but also enhances organizational trust and consumer satisfaction.