Pattern recognition method capable of achieving cluster, classification and metric learning simultaneously

Abstract

The invention relates to a pattern recognition method capable of achieving cluster, classification and metric learning simultaneously. According to the method, the Bayesian theory is used to construct probability relation matrix P between a cluster and a category, and final cluster and category results are enabled to only rely on a cluster center through the matrix; accordingly, cluster learning and classification learning can be achieved simultaneously under a frame. As the statistical relation between the cluster and the category can be reflected through the matrix P, meaningful information can be mined from the P, and design of classifier is transparent. From the cluster point, cluster learning monitoring results are provided, and potential structures of data can be obtained reliably; from the classification point, an effective classification learning mechanism is constructed, and good classification results can be obtained; and from the metric learning point, effective characteristic weight is provided, and significant degree of characteristics can be reflected.

Description

technical field [0001] The invention belongs to the field of pattern recognition, in particular to a pattern recognition method for simultaneously realizing clustering, classification and metric learning. Background technique [0002] Pattern recognition aims to process and analyze the sample data representing things or phenomena to achieve two purposes: to reveal and explain the internal structure of the sample and to judge the category of the sample. According to these two different purposes, traditional pattern recognition machine learning methods can be roughly divided into two categories: clustering learning and classification learning. [0003] Clustering learning uses the similarity between samples to divide samples with the same characteristics into the same meaningful cluster, so as to form a meaningful division of samples. This type of algorithm can discover the potential distribution structure of the sample and better understand and analyze the data, but it canno...

AI Technical Summary

Problems solved by technology

This serial design method makes these algorithms often only emphasize classification learning, and cluster learning is only used as an auxiliary tool for classification learning. Therefore, it is impossible to truly integrate the respective benefits of cluster learning and classification learning under one framework.

In addition, although these methods partially reveal the relationship between the cluster structure and categories, they lack probabilistic meaning and cannot give accurate relevant information

Images