Identification of Anomalous Behavior Through the Observation of an Individual’s Emotional Variation: A Systematic Review

Abstract

Emotion recognition plays a critical role in understanding how individuals perceive and interact with the world by accurately inferring emotions from various sources, such as physical signals and physiological responses. Anomalous emotion recognition, in particular, focuses on identifying emotions that deviate significantly from expected patterns in specific contexts or individuals, including abrupt emotional shifts, inconsistencies across different emotional signals, or persistent, atypical emotional patterns. The importance of recognizing anomalous emotions has grown, particularly in intelligent systems designed for mental health monitoring and crime prevention, where early detection can enhance context-sensitive interventions. This study presents a systematic review that investigates the link between emotional variations and anomalous behaviors across diverse settings. By examining the relationship between emotional shifts and deviant behaviors, the study sheds light on the challenges of detecting emotional anomalies and assesses the efficacy of various recognition methods. A thorough analysis of 102 relevant studies highlights the potential and limitations of current approaches, emphasizing the need for more diverse datasets, improved algorithm robustness, and broader applications to enhance performance in real-world scenarios.

Keywords

• Emotional variations
• anomalous behaviors
• early detection
• systematic literature review
• mental health
• machine learning