Fuzzy division clustering method and device based on deterministic annealing

Abstract

The invention discloses a fuzzy division clustering method and device based on deterministic annealing. According to the invention, the Mahalanobis distance is used as the similarity measure, so thatthe clustering algorithm is suitable for discovering non-spherical clusters; A maximum entropy criterion is used for removing the fuzzy index m, so that the application of the algorithm avoids selecting a fuzzy index m value, and a deterministic annealing mechanism is used for ensuring that the algorithm can obtain a better clustering result under the general condition, thereby well improving theperformance of the clustering algorithm.

Description

technical field

[0001] The invention relates to the field of computer technology, in particular to a method and device for fuzzy partition clustering based on deterministic annealing. Background technique

[0002] With the vigorous development of the network and multimedia, the collected massive text information, image information, video information, audio information and other data make it more and more difficult to manually process these data. The rise of machine learning research has made it possible to process these data through machine learning. In machine learning research, clustering, as an unsupervised learning method, has attracted the attention of researchers from various fields. Data clustering algorithms are widely used in many fields, including machine learning, data mining, pattern recognition, image analysis, and bioinformatics. During the development of clustering algorithms, researchers tried to describe clustering problems from different angles, and propo...

Images