A method for online identification of abnormal state
A technology of abnormal state and identification method, applied in special data processing applications, instruments, electrical digital data processing, etc., can solve the problem that the abnormal point detection method is not suitable for online identification of abnormal state of data flow, etc., to solve the concept transfer and ensure identification. Ability, the effect of reducing consumption
Active Publication Date: 2018-04-24
CHONGQING UNIV OF POSTS & TELECOMM
View PDF2 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] In view of the problems existing in the above background, the present invention provides an online abnormal state identification method to solve the problem that the traditional abnormal point detection method is not suitable for online identification of abnormal states on data streams
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0034] Below, the specific implementation manners of the present invention will be further described in conjunction with the accompanying drawings.
[0035] Such as figure 1 and figure 2 Shown, concrete implementation process and working principle of the present invention are as follows:
[0036] A. Collect data elements in the data stream in real time, obtain a high-dimensional data sample set X containing certain data elements, and preprocess the high-dimensional data sample set X;
[0037] B. Use the outlier factor formula based on the angle distribution to analyze each data element in the set X, so as to obtain the outlier factor value of each data element in the set X;
[0038] C. Divide all data elements in the set X according to the abnormal factor value of each data element, and the set normal set threshold and boundary set threshold. That is to say, data elements are included in the normal set, boundary set and abnormal set. So as to construct the initial normal ...
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
Login to View More Abstract
An online abnormal state identification method is used for real-time online detection of potential abnormal points in high-dimensional data streams. By analyzing the data characteristics of the data stream, a method based on angle distribution is proposed to obtain the outlier factor value corresponding to each data on the data stream. Combined with the need of real-time monitoring of data flow, a small-scale data flow calculation set based on normal set and boundary set is proposed to speed up the operation speed of the abnormal state online identification method. Aiming at the concept transfer problem of big data flow, a real-time update mechanism of normal set and boundary set is proposed to ensure the detection accuracy of abnormal state online recognition method in high-dimensional space. By adopting the method of the present invention, not only can the consumption of time and physical storage be greatly reduced, but also the potential abnormal points in the high-dimensional big data stream can be detected accurately and in real time online, creating conditions for realizing the real-time online evaluation of the data stream, thereby Enhanced the stability of the big data application system.
Description
technical field [0001] The invention relates to technologies such as data mining and abnormal point detection, in particular to an online identification method for abnormal states. Background technique [0002] Outlier detection is one of the most important technical methods in the field of data mining. With the continuous development of science and technology, many practical applications such as e-commerce, network traffic monitoring, wireless communication, logistics and transportation will generate sequential, massive, and rapidly changing infinite data streams. In general, massive data streams are characterized by high dimensionality and concept transfer. Usually, these features greatly hinder anomaly detection on data streams. Therefore, how to effectively mine insecure factors from massive data is a very important topic. [0003] Since the rise of outlier detection research, some famous research institutions and academic units at home and abroad have carried out a l...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): H04L29/06H04L12/24G06F17/30
Inventor 张艳黄质權五景
Owner CHONGQING UNIV OF POSTS & TELECOMM