Abstract:

With the exponential growth of online activities, classifying users based on behavioural risks has become crucial for cybersecurity, fraud prevention, and personalized marketing. According to recent statistics, 82% of data breaches involved some form of human error or online behaviour misuse, and cybercrimes targeting users rose by 50% in the past year alone. Despite the significance, current manual risk detection systems often struggle with real-time analysis and predictive accuracy. In this work, we propose a Behavioural Risk Classifier that classifies users into categories of "Safe," "Neutral," and "Risky" based on their online behaviour

Description:

.