Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A method for correcting abnormal values of a matrix based on a Markov transfer matrix

A transfer matrix and outlier technology, applied to instruments, character and pattern recognition, computer components, etc., can solve problems such as large amount of calculation, low information utilization rate, lack of scene mobility, etc., to achieve small amount of calculation and easy migration sexual effect

Pending Publication Date: 2019-06-18
成都新希望金融信息有限公司
View PDF1 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] For the problems of the above research, the purpose of the present invention is to provide a method for correcting the abnormal value of the matrix based on the Markov transition matrix, to solve the problem of correcting the recognition probability in the prior art, using the hidden Markov transition matrix, The prior information of the hidden Markov transition matrix is ​​not fully utilized, resulting in low information utilization; lack of scene mobility; considering the state transition of two parallel random processes, and the two random processes are related to each other, the problem of large amount of calculation

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 for correcting abnormal values of a matrix based on a Markov transfer matrix
  • A method for correcting abnormal values of a matrix based on a Markov transfer matrix
  • A method for correcting abnormal values of a matrix based on a Markov transfer matrix

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0102] Calculate the recognition probability matrix of the multi-state sequence. The present invention uses the coordinates, speed, acceleration, angular velocity and other data obtained by the mobile phone sensor to identify 20 states of the mobile phone user, and forms a continuous observation time box with a length of 0.5s. A time box obtains the pattern recognition probability in the time box through a pattern recognition model (such as BP neural network, Bayesian decision-making, etc.), and the accuracy of the recognition probability of a single time box depends on the specific pattern recognition model, but the recognition of continuous time boxes When probabilities form a sequence, it is necessary to perform overall outlier identification on the recognition probability of the sequence, and outlier correction after outlier identification. Specifically as figure 1 As shown, the recognition probabilities obtained for each state in each time bin form a time-stamped recognit...

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 method for correcting an abnormal value of a matrix based on a Markov transfer matrix, and belongs to the technical field of data mining and the field of pattern recognition.The method comprises steps of based on a three-connection box abnormal value identification method, carrying out abnormal value judgment on an obtained identification probability matrix, and an abnormal identification probability, namely an abnormal value, is obtained; Customizing a Markov state transition matrix of the identification probability interval; And according to the abnormal value andthe Markov transfer matrix, correcting the abnormal value by using a prior probability and naive Bayes to obtain a corrected recognition probability. The method is used for correcting the abnormal recognition probability.

Description

technical field [0001] The invention discloses a method for correcting the abnormal value of the matrix based on the Markov transition matrix, which is used for correcting the identification probability of the abnormality, and belongs to the field of data mining technology and the field of pattern recognition. Background technique [0002] The accuracy of pattern recognition in pattern recognition is the core issue. Improving the accuracy of pattern recognition can be done from two aspects. One is the accuracy of the algorithm inside the model, and the other is to correct the accuracy from outside the model. [0003] In the prior art, the patent is: A method for indoor pedestrian walking pattern recognition and trajectory tracking CN201310440455.0 discloses that first, according to the characteristics of the terrain in the indoor environment, the indoor walking mode is defined based on the direction and speed; Then use the collected acceleration sensor and geomagnetic sensor...

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): G06K9/62G06K9/00
Inventor 冯诗炀程序段银春刘洪江赵小诣
Owner 成都新希望金融信息有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com