What is it about?
Behaviour analysis has been used widely in anti-fraud transactions. But existing methods of behaviour analysis mainly focus on behaviour patterns, and do not fully consider the behaviour psychology of users in the transaction process, which affects the precise recognition of fraudulent behaviours. It is a difficulty to recognize fraudulent transactions precisely that how to describe the user’s behavioural psychology and integrate the behavioural psychology into the transaction behaviour. So this paper proposes firstly transaction character model based on user’s cautiousness to reflect user’s behavioural psychology. This model is built from the user’s historical normal interaction behaviour data. Then a user behaviour benchmark is established to reflect user’s behaviour pattern from the user’s historical normal transaction behaviour data. To integrate the user’s transaction character and user behaviour, the mapping relationship model is built by using the least squares generalized inverse method. And this model is the core of the fraudulent behaviour recognition method with transaction character. Experiments in fraud detection scenarios show that the new method improved the average recognition performance of four fraud detection indicators (recall rate, precision rate, accuracy rate, F1 value) by 23%. And the method also shows that individual psychological character has a great influence on user behaviour.
Featured Image
Why is it important?
In transaction anti-fraud, the external behavior of users is usually analyzed based on transaction data or interactive behavior. However, the user's external behavior is driven by internal psychological behavior. From the perspective of psychological behavior, this paper innovatively establishes a cautious psychological behavior model by integrating transaction data and interaction behavior data. The model is more effective than the existing anti-fraud methods based on external behavior.
Perspectives
I hope this article will guide researchers to identify and analyze user behavior from the perspective of user psychology. Because the user's internal psychological behavior is more stable than the external behavior performance; It is hoped that psychological behavior research can gain attention not only in the field of anti-fraud, but also in recommender systems, social networks, etc.
Zhaohui Zhang
Donghua University
Read the Original
This page is a summary of: UBRMTC: User Behavior Recognition Model With Transaction Character, IEEE Transactions on Computational Social Systems, January 2023, Institute of Electrical & Electronics Engineers (IEEE),
DOI: 10.1109/tcss.2023.3257227.
You can read the full text:
Contributors
The following have contributed to this page
