Coarse classification method and device based on clustering analysis, terminal equipment and storage medium

Abstract

The invention relates to a coarse classification method based on clustering analysis. Device, terminal equipment and storage medium, The method comprises the following steps of: obtaining a sample; obtaining a first classification result which is determined by classifying the to-be-classified sample data according to a preset clustering algorithm in advance, training the first sample data and thesecond sample data obtained by the first classification to obtain a first-stage SVM classifier, and inputting the to-be-classified sample data of the first-stage SVM classifier into the first-stage SVM classifier for classification; Training by applying the first sample data and the second sample data obtained by last classification to obtain a next-level SVM classifier; and inputting the to-be-classified sample data of the next-level SVM classifier into the next-level SVM classifier, and stopping classification until any one of a first classification stopping condition and a second classification stopping condition in classification stopping conditions is met. Classification error accumulation of each stage of SVM classifier is reduced, and the classifier recognition accuracy in the big data classification process is improved.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a rough classification method, device, terminal equipment and storage medium based on cluster analysis. Background technique [0002] With the development of science and technology, people are faced with a large amount of data in their lives, but they often cannot find the information they need. The information explosion makes how to effectively use and process a large amount of data a common concern in the world today. With the development of database technology, artificial intelligence, mathematical statistics and cloud computing and other technologies, data mining technology has been rapidly applied in all walks of life. [0003] From a technical point of view, data mining is the process of extracting potentially useful information and knowledge hidden in it from a large number of incomplete, noisy, fuzzy, random actual data. From a business point of view, data mining...

AI Technical Summary

Problems solved by technology

[0005] In view of this, the present invention provides a rough classification method, device, terminal equipment and storage medium based on cluster analysis to solve the problem of low recognition accuracy of classifiers in the classification of large data in the prior art

Images