Adaptive PSO-PFCM clustering method for big data mining and clustering

Abstract

The invention discloses an adaptive PSO-PFCM clustering method for big data mining and clustering. Module density calculation and the self-adaptive clustering number of the module density calculationare applied to the big data clustering process, so that the best clustering number can be automatically determined by the big data clustering method which originally needs to manually set clustering data, and the PSO-PFCM algorithm is allowed to be suitable for clustering different original data sets while retaining the excellent performance of the device in the big data processing process, so that a large amount of manual work is reduced.

Description

technical field [0001] The invention belongs to the technical field of big data processing, and in particular relates to an adaptive PSO-PFCM clustering method used in the process of big data mining. Background technique [0002] Clustering Analysis (Clustering Analysis, CA) is an unsupervised machine learning method that studies objects based on mathematical methods and divides given objects. CA is to divide the target object into the corresponding multiple clusters with a certain metric, in order to make the clusters have better similarity, but there are obvious dissimilarities among the clusters. In addition, it can obtain additional useful information from the data without domain knowledge, and can intuitively and quickly reflect the characteristics of the data. However, the traditional fuzzy clustering algorithm needs to set the number of clusters artificially, and different data sets cannot adaptively judge the number of categories, which makes fuzzy clustering of dif...

AI Technical Summary

Problems solved by technology

[0003] Aiming at the above-mentioned deficiencies in the prior art, the adaptive PSO-PFCM clustering method for big data mining clustering provided by the present invention solves

Images