Method and device for building decision-making model

Abstract

The invention relates to a method and device for building a decision-making model. The method comprises the steps of acquiring rule template data, and extracting variable objects and template samples in the rule template data; performing clustering analysis on the variable objects to acquire a clustering result; matching the clustering result with each template sample according to the rule template data, and using the matched clustering result to act as a first feature; respectively calculating the black sample probability of each variable object, and using the black sample probability of each variable object to act as a second feature; and building a decision-making model through the first feature and the second feature. The method and device for building the decision-making model can reduce the dimension and the hierarchy to which the data relates, thereby being conducive to building the decision-making model, and reducing influences imposed on the model performance.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a method and device for constructing a decision model. Background technique [0002] In industries such as insurance and medical care, there are often many document or project reviews, such as preliminary review of insurance company underwriting, bank loan qualification review, medical insurance fraud case review, etc. Most of these document or project reviews rely entirely on manual or complex rules. . Relying on manual review consumes a lot of manpower and time, and complex rules usually involve judging factors with multiple dimensions and complex classification levels. Modeling is difficult and updates are slow, and flexibility is poor, and too many dimensions and levels involved in the data will affect the model. The performance of the enterprise is affected, which is not conducive to business decision-making. Contents of the invention [0003] Based on this, it is nece...

AI Technical Summary

Problems solved by technology

Relying on manual review consumes a lot of manpower and time, and complex rules usually involve judging factors with multiple dimensions and complex classification levels. Modeling is difficult and updates are slow, and flexibility is poor, and too many dimensions and levels involved in the data will affect the model. impact on the performance of the business, which is not conducive to business decisions

Images