Method and model for evaluating network e-commerce loan risk

Abstract

The invention discloses a method and model for evaluating a network e-commerce loan risk. The method comprises the following steps of collecting historical transaction data of a network e-commerce loan customer as a sample set; building a least square twin support vector machine classification model, training the least square twin support vector machine classification model and building a network e-commerce load credit model; predicting default labels of various samples in the sample set according to a built network e-commerce load risk model; comparing a predicted default label with an actual default label for each sample, determining a prediction error of the sample, determining the magnitude of the weight of each sample according to the prediction error; and building the least square twin support vector machine classification model on the basis of the weight of each sample, rebuilding the network e-commerce load risk model and evaluating the load risk. The evaluation accuracy is high.

Description

technical field [0001] The invention relates to a method and model for assessing the risk of online e-commerce lending. Background technique [0002] Internet finance is a brand-new financial service model. It transfers the traditional small loan service to the Internet platform. People who need to borrow can find people who have the ability to lend and are willing to lend based on certain conditions on the online lending platform. It has the advantages of high efficiency, simple operation, oriented to low- and middle-income groups (for borrowers), safety, transparency, and high returns (for lenders), so the online lending platform has been quickly recognized and developed once it was launched [1] . However, compared with traditional loan methods, online lending also has disadvantages. For example, online lending is an unsecured loan, and most of the lenders are ordinary people without professional investment and financial management knowledge. This brings the risk of onli...

AI Technical Summary

Problems solved by technology

The second category is the risk assessment model established using linear data mining algorithms. These models have some common problems: the assessment accuracy is low, and the reason is that the credit data has a nonlinear structure.

However, online e-commerce loan credit data has the characteristics of class imbalance. For example, the normal repayment record of the Paipaidai platform is about 10 times the default record. For such risk data, the first type of error rate of the support vector machine model is relatively low. High, that is, the normal repayment customers are judged as default customers

The same problem exists in the BP neural network, and the BP neural network model has the generalization ability only when the errors of the training set and the prediction set are similar.

Images