Personal credit evaluation method and system based on fusion neural network

Abstract

The invention discloses a personal credit assessment method based on a fusion neural network, and the method comprises the steps: carrying out the data preprocessing: obtaining behavior data for reflecting the personal credit condition, and carrying out the extraction, cleaning and normalization processing; decomposing the different behavior data to obtain multi-layer local feature information; inputting the decomposed behavior data into a first neural network to obtain a prediction data value of the behavior data, and predicting the change trend of the real-time behavior data for assisting inevaluating the personal credit rating; based on the prediction data value obtained by the first neural network, inputting the prediction data value into the second neural network for fitting to obtain the personal credit judgment result. Feature extraction of the behavior data by adopting the LSTM neural network is realized, fitting of multiple behavior data weights is performed through the BP neural network, and the accuracy of personal credit evaluation is improved.

Description

technical field [0001] The invention relates to the technical field of data mining and processing, in particular to a personal credit evaluation method and system based on a fusion neural network. Background technique [0002] Technologies such as the Internet, big data and artificial intelligence are triggering profound changes in the financial industry. With the advent of the era of big data, the amount of data in the financial industry is increasing, the data types are increasing, and the data update speed is accelerating. This is a great opportunity for financial institutions. The personal credit risk assessment work has brought challenges. The traditional personal credit risk assessment methods have been unable to adapt to the changing environment and the development needs of the financial industry. Therefore, it is urgent to build an intelligent credit risk assessment system to help the financial industry Institutions make loan approval decisions. [0003] The feature...

AI Technical Summary

Problems solved by technology

[0002] Technologies such as the Internet, big data and artificial intelligence are triggering profound changes in the financial industry. With the advent of the era of big data, the amount of data in the financial industry is increasing, the data types are increasing, and the data update speed is accelerating. This is a great opportunity for financial institutions. The personal credit risk assessment work has brought challenges. The traditional personal credit risk assessment methods have been unable to adapt to the changing environment and the development needs of the financial industry. Therefore, it is urgent to build an intelligent credit risk assessment system to help the financial industry Institutions make loan approval decisions

[0003] The feature data used to evaluate personal credit in the prior art is mostly static information. Modeling is performed based on various feature data, and the weight distribution of various feature data is fitted to obtain personal credit evaluation results. However, static information data such as personal features , occupation information, family information, education level, etc. will not change in the short term, and cannot reflect personal income fluctuations and credit fluctuations. There are few studies on the dynamic evaluation of personal credit

Images