A real estate collateral risk control method, system and storage medium

Abstract

The present invention establishes an intelligent valuation algorithm system on the basis of collateral big data. The valuation process adopts a dual-network design. The first network is used to judge the most suitable evaluation algorithm for the collateral and increase the weight of the algorithm's valuation; the second network is used to collect user feedback information, whether it is correct valuation or biased On the one hand, the computer automatically analyzes and records, on the other hand, human participation helps the computer to correct the deviation or error, similar to a deviation or error memory stick, all of which are fed back to the first network. The two networks are constantly running in a loop, learning from the right and learning from the wrong. Over time, experience will be formed and continuous improvement. It can be concluded that the accuracy rate will become higher and higher, so as to realize automatic and accurate valuation of collateral in batches, and in the process of accurate valuation Based on risk analysis.

Description

technical field [0001] The present invention relates to a collateral risk control method and system, in particular to a real estate collateral risk control method and system. Background technique [0002] At present, the traditional method of bank financial risk control is to manually review the collateral on the spot. With the increase of tens of thousands of business volume, the time consumption has also increased dramatically, and the technology still stays at the traditional stage of relying on the experience and judgment of appraisers. Sensation, experience, and relational decision-making dominate, leading to inconsistencies in review and judgment standards. [0003] There are two ways to assess the value of collateral: internal assessment and external assessment, and some of them use a combination of self-assessment and external assessment. But there are following problems: [0004] At present, most domestic banks only focus on the value of a single credit project, a...

AI Technical Summary

Problems solved by technology

[0006] After the traditional method uses manual valuation, it is necessary to manually save the valuation results digitally, which is prone to errors, and the valuation process is susceptible to manual interference.

[0007] The cost of manual valuation in traditional methods is high, especially in large quantities

[0008] The traditional method uses a fixed algorithm for the valuation of real estate. Using the same algorithm for different types of real estate can easily lead to inaccurate valuation results

Images