A method of user abnormal behavior detection based on neural network clustering

A neural network and detection method technology, applied in the field of abnormal user behavior detection based on neural network clustering, can solve the problem of high false alarm rate, achieve the effect of improving decomposition speed and signal-to-noise ratio

Active Publication Date: 2018-04-13
HENAN UNIV OF SCI & TECH
View PDF2 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But this technology also has disadvantages, its false alarm rate is relatively high

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
  • A method of user abnormal behavior detection based on neural network clustering
  • A method of user abnormal behavior detection based on neural network clustering
  • A method of user abnormal behavior detection based on neural network clustering

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] As shown in the figure, a method for detecting abnormal user behavior based on neural network clustering includes the following steps:

[0032] Step 1. Assuming that each user is an object, the behavior data of the object is used as the attribute of the object, and the matrix real matrix A of n*m dimensions is expressed as an object-attribute structure, and the matrix A is expressed as Among them, X ij Expressed as the j-th attribute of the i-th object; step 2, using the matrix A in step 1 with the formula A=USV T Carry out SVD decomposition, where, U=(u 1 , u 2 ,...,u m ), V=(v 1 ,v 2 ,...,v n ), S=diag(δ 1 ,δ 2 ,...δ r ),δ i >0(i=1,2,...,r), r=rank(A), lambda i 2 (i=1,2,...,r), λ i for A T All the non-zero eigenvalues ​​of A, arranged in descending order, are the singular values ​​of A;

[0033] Step 3. Calculate the order of the effective rank of the decomposed SVD matrix according to the following formulas (1)-(3) to denoise the user behavior infor...

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 user abnormal behavior detection method based on neural network clustering. First, the user's behavior data set matrix is ​​decomposed by SVD and denoised, and then the denoised matrix is ​​input into the input layer of the neural network, and then the denoised matrix is ​​input into the input layer of the neural network. The hidden layer of the network carries out the weights of all attributes of each user and standardizes them, outputs them in the output layer, and finally calculates the similarity between each user and each user in the normal behavior model database and the size of the threshold, if greater than the threshold Directly for abnormal behavior and carry out corresponding prompts and preventive measures, otherwise it is normal behavior, merge it into the normal model database to update the database in real time; the present invention effectively achieves high detection rate and low false alarm through the cooperation of various parts Rate.

Description

technical field [0001] The invention relates to the technical field of information security credibility, in particular to a method for detecting user abnormal behavior based on neural network clustering. Background technique [0002] Today is in the era of big data, all kinds of user information are stored in the computer, which brings convenience to people's work and life, but also makes information security face a severe test. With the popularization of computers, various hackers and intrusions emerge in endlessly , with the network attack technology becoming more mature and changing, the traditional passive defense methods obviously cannot solve the security problem of user information well. For various passive defense measures, people are more inclined to active detection technology, and adopt Graph clustering is a popular method for detecting abnormal user behavior. [0003] Using the method of graph clustering to realize the identification and detection of user abnorm...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G06F21/56G06K9/62
CPCG06F21/566G06F18/24
Inventor 郑瑞娟张明川吴庆涛白秀玲魏汪洋赵海霞陈京张童王成磊杨丽
Owner HENAN UNIV OF SCI & TECH
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