While we often focus on the literal content of online communication, the underlying ”mood” or disposition of a user can be just as crucial in understanding their behaviour. The ambitious goal here is to utilise a ”window-based logistic regression-controlled environment” to predict benign and malicious moods. For the model to function efficiently, we ere able to reduce the infinite spectrum of human emotion into two statistically separable classes: “Benign”😌 and “Hacker”😈 mood, where sixty (60) participants were given a website to interact with in a freewill and choice of surfing. The first state represents calm, focused, and productive behaviour, while the latter state is characterized by high cognitive overload, frustration, and rapid context. The results show that for benign cases, the eye patterns of users typically exhibited stable, focused fixations on relevant content areas, which produce calm emojis, while for the hacker mood, Individuals simulating malicious intent often displayed starkly different patterns represented as rapid, disjointed saccades that predict hacker emojis. The accuracy of our model in distinguishing between benign and hacker moods via eye movements was remarkably high and indicates the authenticity of the model.

Keywords: Mood disposition, Fixation, Saccade, Eye patterns, Emojis Benign mood, Hacker mood, Logistic Regression