Predictive Modeling of Emotional and Behavioral Patterns in Cluster B Users Using AI Web Viewers on Social Media

Abstract: The integration of artificial intelligence (AI) with psychological profiling has opened new frontiers in understanding human emotions and behaviors within digital environments. This study proposes a predictive modeling framework designed to analyze emotional and behavioral patterns in users exhibiting Cluster B personality traits—comprising narcissistic, histrionic, borderline, and antisocial tendencies—across major social media platforms. Using AI-driven web viewers and behavioral analytics, data were collected from 1,200 verified social media accounts across Twitter (now X), Instagram, and Reddit between 2020 and 2024. Emotional valence, sentiment polarity, posting frequency, linguistic tone, and engagement ratios were extracted and analyzed through supervised learning algorithms including Random Forest, Support Vector Machines (SVM), and Recurrent Neural Networks (RNN). The results demonstrate significant predictive correlations between linguistic sentiment and behavioral impulsivity among Cluster B–type users, with AI models achieving up to 89.7% accuracy in predicting mood-driven posting patterns. Comparative evaluations indicate that RNN-based models outperform traditional regression methods in detecting fluctuations in emotional intensity and social interaction frequency. Additionally, AI web viewers effectively identify cyclical emotional dynamics aligned with online validation-seeking behavior. These insights underline the potential of AI-assisted psychological analytics in early identification and intervention for maladaptive online behaviors. This research bridges computational psychology and data-driven behavioral science by presenting a scalable, ethically grounded framework for analyzing mental health dynamics in virtual social ecosystems. The implications extend to digital mental health monitoring, AI ethics, and the development of responsible algorithms capable of understanding human affectivity.

 Keywords: Artificial Intelligence, Cluster B Personality, Emotional Behavior Prediction, Social Media Analytics, Psychological Profiling