Credit fraud detection method and system

A detection method and detection system technology, applied in the credit field, can solve the problem of incompatibility between accuracy and training speed, and achieve the effects of simple structure, saving computing resources, and improving training speed and classification speed.

Pending Publication Date: 2020-06-16
BEIJING INFORMATION SCI & TECH UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a credit fraud detection method and system, which solves the

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Credit fraud detection method and system
  • Credit fraud detection method and system
  • Credit fraud detection method and system

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0107] This embodiment provides a specific implementation:

[0108] Obtain credit sample data. The credit sample data is the transaction data of European credit card holders in September 2013. The credit sample data comes from the research group (Worldline and the Machine Learning) located at the Université Libre de Bruxelles (ULB) in Belgium. Group).

[0109] PCA transformation is used to reduce the dimension of the credit sample data to obtain the dimension-reduced credit sample data. The dimension-reduced credit sample data contains 284,807 transaction records, and each transaction record contains 31 fields. For details, see Table 1. Each row of record fields A feature of:

[0110] Table 1 The specific information of a transaction record

[0111]

[0112]

[0113] In Table 1, No. indicates the number, Time indicates the time, Float indicates that the data type is a floating point type, V1-V28 indicates the attribute name, Amount indicates the amount, and Class indic...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a credit fraud detection method and system, and relates to the field of credit. The method comprises the steps of obtaining credit sample data, dividing the credit sample datainto a training set and a test set, training a feature data window and a logistic regression model by using the training set to obtain a credit fraud detection initial model, debugging the credit fraud detection initial model by using the test set to obtain a credit fraud detection model, obtaining credit data to be tested, inputting the credit data to be detected into the credit fraud detection model to obtain a logistic regression classification result, and performing decision fusion on the logistic regression classification result to obtain a credit fraud detection result. According to thecredit fraud detection method and system, feature scanning and logistic regression are combined, feature scanning and logistic regression are performed on the to-be-detected credit data, and finally decision fusion is performed, so that the accuracy of credit fraud detection is improved, the credit fraud detection model is simple in structure, and the training speed and the classification speed are improved.

Description

technical field [0001] The invention relates to the field of credit, in particular to a credit fraud detection method and system. Background technique [0002] At present, there are two main methods for credit fraud detection: anti-fraud methods based on rule engines and anti-fraud methods based on machine learning. However, the rule engine relies heavily on the experience and lessons of experts, and the effect of the rule system formulated by different decision makers is often quite different, so the rule engine has great limitations. Machine learning has a wide range of applications in the field of anti-fraud, and classification algorithms such as logistic regression, support vector machines, and random forests are the most common; in addition, algorithms such as frequent itemset mining and neural networks also have good results in the field of anti-fraud. However, for models with simple structures such as linear regression and logistic regression, the accuracy of the mod...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06Q40/02G06K9/62
CPCG06Q40/03G06F18/241G06F18/214
Inventor 康海燕张浩
Owner BEIJING INFORMATION SCI & TECH UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap