Sample classification method and apparatus

Abstract

The invention discloses a sample classification method and apparatus, and belongs to the technical field of computers. The sample classification method includes the steps: calculating a similarity between a test sample and the cluster center of a plurality of sub-clusters, and determining the selection interval according to the similarity and the preset threshold, wherein the sub-clusters are obtained through clustering of a training sample set; selecting training samples having the similarity with the clustering center in the selection interval from the sub-clusters corresponding to the clustering center with the highest similarity; and taking the selected training samples as a new training sample set, and classifying the test samples. For each test sample, the sample classification method selects the training samples from the sub-clusters corresponding to the clustering center with the highest similarity according to the determined selection interval, and classifies each test sampleby using the selected training samples, thus reducing the number of subsequent training samples to be classified, and improving the sample classification efficiency in a big data environment.

Description

technical field [0001] The invention relates to the field of computers, in particular to a sample classification method and device. Background technique [0002] Because of its simplicity and ease of implementation, the K-nearest neighbor algorithm is widely used in many fields, such as face recognition, gene classification, decision support, etc. The basic idea of ​​the K-nearest neighbor algorithm is: for a given test sample x, find its K nearest neighbor samples in the training sample set, and determine the category of the test sample x according to the categories of the K nearest neighbor samples. [0003] In the process of realizing the present invention, the inventor finds that there are at least the following problems in the prior art: the K nearest neighbor algorithm needs to calculate the distance of each training sample in the test sample and the training sample set one by one (or Similarity), when the training sample set is large data, the above calculation proce...

AI Technical Summary

Problems solved by technology

[0003] In the process of realizing the present invention, the inventor finds that there are at least the following problems in the prior art: the K nearest neighbor algorithm needs to calculate the distance of each training sample in the test sample and the training sample set one by one (or Similarity), when the training sample set is large data, the above calculation process will generate high overhead, resulting in very low efficiency of the algorithm, or even infeasible

Images