Method for classifying N central points based on gene expression programming

Abstract

The invention discloses a method for classifying N central points based on gene expression programming and belongs to the technical field of machine learning. The method includes the steps of firstly, partitioning to-be-classified training data into a classification-marker-contained data set and a classification-marker-free data set; secondly, searching classification central points from a multi-dimensional space of the classification-marker-free data set according to a distance formula; thirdly, assessing accuracy of the searched classification central points on the classification-marker-contained data set; fourthly, using a gene expression programming method for expressing, searching and updating the classification central points to obtain new classification central points; fifthly, classifying unknown points according to the distances between the unknown points and the new classification central points; sixthly, repeating the third step, the fourth step and the fifth step until an objective function satisfies stop conditions. By the aid of the method for classifying the N central points based on gene expression programming, calculated quantity can be decreased while susceptibility to unbalanced data is avoided.

Description

technical field

[0001] The invention relates to an N-centroid classification method based on gene expression programming, which belongs to the technical field of machine learning. Background technique

[0002] Distance-based classification methods represent each datum as a numeric vector, constructing a class center for each class. When classifying, calculate the distance between the data to be classified and each class center point, and the category of the data to be classified is the class represented by the nearest class center point.

[0003] The k-nearest neighbor classification method is a commonly used distance-based classification method. It avoids the difficulty of directly constructing the class center, and finds k data points closest to the data to be classified by calculation, and the category of the data to be classified is the category with the largest number of k data. The k-nearest neighbor method is a commonly used distance-based classification method, but...

Images