Adaptive Fraud Detection Feature Acquisition for Faster Screening
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fraud detection systems require a large number of feature amounts for all users, leading to prolonged detection times.
Innovation Solution
A fraud detection system that varies the acquisition time and depth of feature amounts based on an overall score, performing a simple check for low scores and a careful check for high scores, thereby reducing the overall detection time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If common settings are used for feature amount acquisition for all users, then the system maintains uniform processing standards, but the fraud detection time becomes excessively long
Solution Approach 1:
The patent applies local quality by customizing feature amount acquisition parameters according to each user's fraud level. Users with lower fraud levels have their feature amounts acquired with shorter acquisition times and fewer features, while users with higher fraud levels undergo more comprehensive feature acquisition. This resolves the contradiction by making the detection process adaptive to individual user risk profiles rather than applying uniform standards to all users.
Solution Approach 2:
The patent implements dynamics by making the feature amount acquisition method dynamic and adjustable based on fraud level assessments. The system determines acquisition methods such as acquisition time, number of features, and feature types dynamically according to each user's risk score. This allows the system to balance detection accuracy and speed by adapting the acquisition parameters in real-time based on user-specific conditions.
2Measurement precision
If a large number of feature amounts are acquired for all users, then detection accuracy is improved, but processing load increases and efficiency decreases
Solution Approach 1:
The patent applies partial action by acquiring only the necessary number of feature amounts for each user based on their fraud level. Instead of acquiring all possible features for every user, the system determines an optimal subset of features to acquire for each user. This reduces the overall processing load while maintaining sufficient detection accuracy by focusing computational resources on users who need more thorough analysis.
Solution Approach 2:
The patent implements parameter changes by adjusting feature amount acquisition parameters such as the number of features, acquisition time, and feature types based on user-specific fraud levels. The system changes these parameters dynamically to optimize the balance between detection accuracy and processing efficiency, rather than using fixed parameters for all users.
3Ease of operation
If feature acquisition time is shortened for users with lower fraud levels, then user convenience is improved, but the number of requests and overall acquisition time management becomes more complex
Solution Approach 1:
The patent applies feedback by using the fraud level assessment results to automatically adjust feature acquisition parameters. The system receives feedback from the fraud level determination and uses this information to control the acquisition process, creating a closed-loop system that automatically optimizes acquisition parameters based on user risk profiles without requiring manual intervention.
Solution Approach 2:
The patent implements self-service by enabling the system to automatically determine and execute appropriate acquisition methods based on fraud level assessments. The system serves itself by autonomously adjusting acquisition parameters, managing the complexity internally without requiring external control, thereby simplifying the user experience while handling the computational complexity behind the scenes.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Score acquisition means (103) of a fraud detection system (S) is configured to acquire, based on an action performed by each of a plurality of users, a score relating to a fraud level of the user. Determination means (104) is configured to determine, based on the score of each of the plurality of users, an acquisition method for a feature amount of the user such that an acquisition time of the feature amount becomes shorter as the fraud level becomes lower. Feature amount acquisition means (105) is configured to acquire the feature amount of each of the plurality of users based on the acquisition method determined for the user. Detection means (106) is configured to detect fraud made by each of the plurality of users based on the feature amount of the user.