Constructing an Investment Scam Detection Model Based on Emotional Fluctuations Throughout the Investment Scam Life Cycle
Chin Kuo, Seng‐Su Tsang · Deviant Behavior · 2023
In recent years, the COVID-19 pandemic has increased investment scams as individuals seek alternative investment opportunities. This study employs emotion analysis to examine the emotional expressions of scammers in investment scams and proposes an investment scam detection model to identify fraudulent behavior. Using data from seven real-world investment scam chat rooms, we perform a two-stage analysis to evaluate the emotional fluctuations of scammers and construct a machine learning-based investment scam detection model. Our findings reveal distinct patterns of emotional fluctuation in investment scam chat rooms according to the established Investment Scam Life Cycle (ISLC), which portrays the progression of fraudulent behavior. We confirm the applicability of machine learning in scam detection, and the emotion analysis presented in this study can assist researchers and fraud examiners in identifying scam scenarios and reducing financial losses.