用户画像生成方法及装置
By considering the full lifecycle information of target users in user profile generation and using neural networks to generate stable user profiles, the problem of declining model stability in existing technologies is solved, and the accuracy of credit risk assessment is improved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- PING AN BANK CO LTD
- Filing Date
- 2023-03-29
- Publication Date
- 2026-07-17
AI Technical Summary
In existing technologies, enterprise user profile generation models suffer from decreased model stability and reduced accuracy in credit risk assessment due to discrepancies between training samples and actual user information applying for financial services.
By acquiring target users' asset information before applying for financial loans, their transaction behavior information during the loan application process, and their asset information after the loan application, and inputting this information into a trained neural network to generate user profiles, the model takes into account the information of all users, avoids training network bias, and improves model stability.
It improved the accuracy of user profiling and increased the accuracy of identifying potential credit risks for corporate users.
Smart Images
Figure CN116383720B_ABST