Method for selecting characteristic facing to complicated mode classification

Abstract

The invention provides a feature selection method for complex pattern classification. On the basis of the traditional genetic method, the method allows a species group to be divided into a plurality of sub-species groups by improving the structure of the species group into a dicyclic intelligent agent network configuration. Information is passed among the sub-species groups through the sharing intelligent agent. The genetic evolution of all sub-species groups is conducted simultaneously. The improved dynamic neighborhood competitive operation and the neighborhood adaptive crossover operating method improve the efficiency of the genetic evolution. At the same time, binary coding mode is introduced to express whether a certain feature is selected or not, thereby facilitating coding and decoding, and achieving the high efficient feature selection. Compared with the traditional feature selection, the feature selection method has the advantages of higher adaptability, and quick search in feature space with high dimension and multimodal, thereby effectively avoiding being trapped in local extremum and acquiring relatively satisfied feature selection results.

Description

technical field [0001] The invention belongs to the technical field of pattern classification, and in particular relates to a feature selection method for complex pattern classification. Background technique [0002] Pattern classification is currently widely used in various fields of society, such as image classification, data mining, information retrieval, information extraction, speech recognition, etc., and its processing methods usually include the following parts: sample preprocessing, feature extraction, feature selection, classification . Among them, feature selection is an important preprocessing process in pattern classification systems. In the pattern classification system, samples after feature extraction often contain a large number of features, and feature selection is to filter out features that are irrelevant or less useful for classification from these large number of features, and select features that are very useful for classification. device classificat...

AI Technical Summary

Problems solved by technology

But these improved genetic methods mainly focus on the improvement of genetic evolution manipulation [3-5] , rarely involves the improvement of the population structure; secondly, the genetic operation must be carried out in the entire population, which not only requires a large amount of calculation (for example, the selection process needs to select individuals in the entire population), but also easily reduces the diversity of the population, which is not realistic. The state of evolution in nature, leading to premature convergence; therefore, improved genetic methods for high performance are yet to be investigated

Weicai Zhong et al. introduced agents to realize a new population network structure-a grid-like agent structure, and combined with the search method of genetic methods, proposed a multi-agent genetic method, which has made important improvements in the field of numerical optimization effect, but it has not been introduced into feature selection for research, and its optimization speed and accuracy need to be improved [6]

Moreover, we found in the research that in the agent genetic method, the neighborhood competition of agent individuals is not dynamic, and because it adopts the network structure of four-neighborhood agents, it is easy to cause some dominant individuals to occupy multiple network grid points, that is, easy Lead to local top dominance, which is not conducive to maintaining the diversity of the population, and is prone to "premature" phenomenon

Images