Multi-target evolutionary fuzzy rule classification method based on decomposition

A technology of multi-objective evolution and classification method, which is applied in the field of classification in data mining, and can solve problems such as difficult identification of minority classes

Active Publication Date: 2015-07-29
XIDIAN UNIV
View PDF2 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to propose a decomposition-based method with high recognition accuracy and fast operation speed for the problem that the minority class is not easy to be r

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
  • Multi-target evolutionary fuzzy rule classification method based on decomposition
  • Multi-target evolutionary fuzzy rule classification method based on decomposition
  • Multi-target evolutionary fuzzy rule classification method based on decomposition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0051]Example 1: In recent years, with the continuous development of science and technology, people have a deeper understanding of the living space, and more and more problems encountered, more and more complex, in which the data classification problem is especially unbalanced data classification The problem is becoming more and more prominent in front of us. Unbalanced data classification is widely used in people's production and life, such as biomedical recognition, to obtain a certain data from yeast cells and determine whether the data comes from a certain internal part of yeast cells , the yeast cell has a total of 10 internal parts, the part to which the acquired data belongs has less data, and the remaining 9 internal parts have more data, which causes the imbalance between data, tumor In the detection, it is determined whether a person has breast cancer. In the crowd, the number of people who really have breast cancer is small, while the number of normal people without...

Embodiment 2

[0063] Embodiment 2: The multi-objective evolutionary fuzzy rule classification method based on decomposition is the same as embodiment 1, wherein step 4 initializes the population P formed by pop fuzzy classifiers and uses the fuzzy rule weight formula with weighting factors to determine the fuzzy rule weight w i ,i∈{1,...,pop}, in this example, initialize a population P consisting of 150 fuzzy classifiers and use the fuzzy rule weight formula with weighting factors to determine the fuzzy rule weight w i ,i∈{1,...,150} includes the following steps:

[0064] 4a. Randomly select a piece of data x from the training data set X rand =[x rand,1 ,...,x rand,n ,y rand ], corresponding to the piece of data x rand There is a fuzzy rule r rand =[rrand,1 ,...,r rand,n ,w rand , l rand ], where the fuzzy rule r rand The attribute values ​​of the first n items are respectively recorded with the data x rand The fuzzy partition labels corresponding to the n attributes of , w rand ...

Embodiment 3

[0082] Embodiment 3: The decomposition-based multi-objective evolutionary fuzzy rule classification method is the same as that in Embodiment 1-2, wherein step 7 performs an evolutionary operation on the original population P and adopts the Chebyshev update method to update each chromosome in turn, and the result is composed of pop chromosomes The evolutionary population P′ of , in this example, the evolutionary population P′ composed of 150 chromosomes is obtained, including the following steps:

[0083] 7a. Chromosome Chromosome h ,h∈{1,...,pop} performs a single-point crossover operation to obtain two crossover daughter chromosomes chro h,cr_1 , chro h,cr_2 ,h∈{1,...,pop}, in this example, for chromosome chro h, h∈{1,...,150} performs a single-point crossover operation to obtain two crossover daughter chromosomes chro h,cr_1 , chro h,cr_2 ,h∈{1,...,150}.

[0084] 7b. Chromosomes of two crossed progeny h,cr_1 , chro h,cr_2 ,h∈{1,...,pop} is used as the mutated parent c...

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 multi-target evolutionary fuzzy rule classification method based on decomposition, which mainly solves the problem of poor classification effect of an existing classification method on unbalanced data. The multi-target evolutionary fuzzy rule classification method comprises the steps of: obtaining a training data set and a test data set; normalizing and dividing the training data set into a majority class and a minority class; initializing an ignoring probability, a fuzzy partition number and a membership degree function; initializing an original group, and determining weight by adopting a fuzzy rule weight formula with a weighting factor; determining stopping criteria for iteration, iteration times, a step size and an ideal point; dividing direction vectors according to groups; performing evolutionary operation on the original group, and updating the original group by adopting a Chebyshev update mode until the criteria for iteration is stopped; obtaining classification results of the test data set; then projecting to obtain AUCH and output. The multi-target evolutionary fuzzy rule classification method has the advantages of high operating speed and good classification effect and can be applied in the technical fields of tumor detection, error detection, credit card fraud detection, spam messages recognition and the like.

Description

technical field [0001] The invention belongs to the technical field of classification in data mining, in particular to a method for optimizing a fuzzy classifier by using an evolutionary algorithm in the field of unbalanced data classification. Specifically, it is a multi-objective evolutionary fuzzy rule classification method based on decomposition. It is mainly used for the classification of unbalanced data in the fields of biomedical identification, tumor detection, credit card fraud detection, and spam identification. Background technique [0002] With the continuous advancement of science and technology, data mining technology is more and more widely used in our production and life, and as an important branch of the data mining field, classification technology is also more and more valued by people. Classification technology refers to the technology of using known categories of data to train classification models, and then using this classification model to predict cat...

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): G06K9/62G06N3/12
CPCG06N3/12G06F18/24
Inventor 刘若辰焦李成宋晓林马晨琳于昕王爽马晶晶刘红英
Owner XIDIAN 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