Subsistence allowance identification method based on a knowledge graph

Abstract

The invention provides a subsistence allowance identification method based on a knowledge graph. The subsistence allowance identification method comprises the following steps: 1, acquiring behavior and attribute information data of a person; 2, constructing relationships betweencharacters and characters characters and attributes, and characters and commodity entities to form a character knowledgegraph; 3, counting character feature data by searching for the character knowledge graph; 4, constructing a training data set according to the counted character feature data; 5, establishing a classification model, and training a decision tree classifier through the training data set; 6, calculating and judging the character feature data by using the trained decision tree classifier, judging whether the character corresponding to the character feature data is a person enjoying the subsistence allowance or not, and completing classification of the person enjoying the subsistence allowance and aperson without the subsistence allowance; and 7, if more than N% of the predicted people are subsistence allowance fraud personnel, the subsistence allowance system may have vulnerabilities, and adjustment needs to be performed. According to the invention, suggestions can be provided for decision making and policy implementation of related departments, so that the life of common people is guaranteed.

Description

Technical field [0001] The invention relates to the field of communication technology, in particular to a low-guarantee identification method based on a knowledge graph. Background technique [0002] The act of defrauding the subsistence allowance refers to the situation where the minimum subsistence allowance is not met or the subsistence allowance is not granted, but the situation is still received. However, the current behavior of defrauding Dibao is caused by the concealment of information, and the asymmetry of information has led to loopholes. Both explicit personal information and hidden personal information are important information for identifying fraudulent insurance. There are relatively few explicit personal information such as real income and consumption information that can be obtained, while it is more difficult to obtain hidden personal information . At present, the source of identification data on insurance fraud is not comprehensive enough. It is mainly judged ...

AI Technical Summary

Problems solved by technology

At present, the source of identification data for fraudulent insurance behavior is not comprehensive enough. It is mainly judged by associating subsistence allowance information and social security information, while more hidden personal information and behaviors are still unidentifiable, and the reason for the inability to identify is largely due to the review of personnel. due to incomplete information

Images