Distributed multi-kernel decision-making system and method for credit granting industry

Abstract

The invention discloses a distributed multi-kernel decision-making system and method for the credit granting industry. The method comprises the following steps: 1, building an integral decision-making network, then constructing different decision-making nodes, and storing the decision-making nodes on a server, which can be stored in different servers; 2, calling different decision nodes on the server by the system by using an RPC (Remote Procedure Call), and summarizing the results in a result output module; and 3, obtaining a final decision result. According to the system provided by the invention, a plurality of decision nodes can be executed in parallel, and the decision-making speed is increased. In addition, the invention also defines a rule updating language. The service personnel can update the rules of the existing decision-making system by using the language, and the daily use of the system is greatly facilitated.

Description

technical field [0001] The invention relates to the field of application development in the credit industry, in particular to a distributed multi-core decision-making system and method for the credit industry in the field requiring a large amount of transaction processing. Background technique [0002] The decision engine is one of the cores of financial technology. It automatically judges whether the current operation should be executed by using a series of rules. For example, when making a credit decision, the bank needs to judge whether it should give the current user a loan based on information such as the current user's credit status and work income, and how much the loan is more appropriate. The decision engine contains the following values. First, it allows business departments to quickly define, update and maintain business rules, scoring models and decision-making processes; second, it allows business analysts to quickly deploy and implement business rules and decis...

AI Technical Summary

Problems solved by technology

From the client side, most of the existing decision-making system clients can only be used to view the predicted results, and cannot effectively edit the existing rules in the model

From the server side, there are also several problems: First, the existing decision-making system may still use the method of hard-coding rules in the code, which is inconvenient for future expansion and management

The second is that some systems that use the Rete network as a decision engine may have a complex Alpha network due to the complexity of the scene and take up a lot of memory space.

Third, the rules in the decision engine are generally executed sequentially, so when a rule is very complex, the time spent in decision-making may be very large

To sum up, the current decision-making system in the credit industry cannot bear the increasing business requirements

Images