Index processing implementation method applied to decision engine

Abstract

The invention discloses an index processing implementation method applied to a decision engine, and the method comprises the following steps: S1, obtaining index data, preprocessing the index data, and inputting the preprocessed index data into an index dictionary module; S2, performing index definition in the index dictionary module; S3, performing feature index extraction on the defined indexes to generate feature index values; S4, converting the feature indexes into index processing json scripts, and generating business indexes; and S5, binding the business indexes to the rule model or the score card model for decision rule operation in the decision engine. After the feature index processing calculation is completed, the corresponding feature index value is stored in the index library, and when the decision engine rule model runs, the business index value is subjected to index processing calculation based on the feature index value in the index library, so that the calculation process of the feature index value is omitted, the difficulty of using the decision engine index by business personnel is reduced, and business personnel can quickly adjust the indexes according to business requirements.

Description

technical field [0001] The invention relates to the technical field of decision engines, in particular to an index processing implementation method applied in decision engines. Background technique [0002] The decision-making engine has been widely used in the financial industry. As the core of financial technology, it needs to support the automatic processing of a large amount of business data, and output the final decision-making results according to the decision-making rule model, which can greatly liberate the bottleneck and efficiency of manual processing; The core idea is to extract the user's identity, behavior and data characteristics based on a large number of real user sample data, and to eliminate them through algorithm technology from the perspective of probability, so as to ensure that the overdue probability of the user group that passes the decision-making rule is at a relative lower range. The characteristic index value obtained by processing and refining t...

AI Technical Summary

Problems solved by technology

[0006] The present invention mainly solves the problem that the index processing is difficult to adjust due to the high complexity of index processing in the existing technology; it provides an index processing implementation method applied to the decision-making engine, which reduces the difficulty for business personnel to use the decision-making engine index, so that business personnel can Demand Quick Adjustment Indicator

Images