Internet credit evaluation method and system

Abstract

The invention discloses an Internet credit evaluation method and system. According to the method, after a training data set is acquired, weights can be respectively set for each training sample in thetraining data set according to the preset strategy, secondly, a preset evaluation model is trained through utilizing the weighted training data set to acquire a post-training evaluation model, basedon the post-training evaluation model, the user Internet credit is evaluated. The Internet credit evaluation method is advantaged in that rationality and accuracy of evaluation can be greatly improved, and the application performance is improved.

Description

technical field

[0001] The invention relates to the field of communication technology, in particular to an Internet credit evaluation method and system. Background technique

[0002] With the advent of the era of big data, the application of Internet credit information is becoming more and more extensive. In addition to being applied to Internet finance, it can also cover other life scenarios, such as taxis, car rentals, or hotel reservations. Therefore, How to ensure the accuracy and fairness of Internet credit evaluation has gradually become a concern of people.

[0003] In the existing technology, it is generally possible to collect user behavior data during the training period as a training data set, and then extract user features from it, and use machine learning algorithms such as decision trees and logistic regression to establish a credit scoring model, and based on The credit scoring model evaluates the user's creditworthiness. Among them, the training data set is...

Images