Research and Analysis of Potential Correlation of Telecommunications Fraud Data Based on Big Five Personalities

Abstract In recent years, telecommunications fraud cases show a high incidence, more and more scholars have studied how to stop telecommunications fraud cases from the root cause. According to psychological knowledge, we can know that personality traits can affect a person's social behavior. Therefore, this paper proposes a personality trait recognition model based on the Big Five personality theory combining Bert and SVM to study how to block the occurrence of telecommunications fraud cases from the root. Compared with the previous model, the average accuracy of the recognition of personality traits was improved by 0.3%, reaching 61.2%. At the same time, combined with the telecommunications fraud data in a certain place, it is analyzed that in telecommunications fraud, there is a correlation between the probability of being defrauded, the means of being defrauded, the location where the fraud occurs and personality characteristics. Our research results may provide reference for the government and relevant departments to manage telecommunications fraud.