Clustering method based on adjacent grid search

Abstract

The invention discloses a clustering method based on an adjacent grid search strategy, and the method comprises the steps: firstly carrying out grid division of original data: enabling an original data set to be divided into a limited number of cells through a multi-dimensional space grid, and carrying out the denoising processing when necessary; secondly, performing grid clustering on the divided data: processing the denoised grids by utilizing a halo threshold value, and dividing the denoised grids into halo cell elements and core cell elements; establishing an adjacent grid operator for quickly searching an adjacent cell element of one cell element; achieving a clustering process through two steps of core cell element clustering and halo cell element division, dividing all core cell elements into a plurality of class clusters through a traversal algorithm, and dividing halo cell elements into existing class clusters based on cell element distances; and finally, clustering optimization is carried out according to data characteristics and user requirements. Compared with the prior art, the method has the advantages that a new clustering method can be provided for rapidly increasing the dimension of the sample set, and class clusters with complex boundary shapes can be effectively identified.

Description

technical field [0001] The invention relates to the technical field of unsupervised pattern recognition and data mining, in particular to a grid-based clustering method. Background technique [0002] With the development of big data and network technology, there is a large amount of data surplus in various disciplines and fields, so cluster analysis has become an increasingly important technology. The process of dividing a collection of physical or abstract objects into classes of similar objects is called clustering. A cluster generated by clustering is a collection of data objects that are similar to objects in the same cluster and different from objects in other clusters. With the application of clustering in various fields, higher robustness requirements are put forward for clustering algorithms. The following specific datasets are of increasing interest in many applications: (1) noisy datasets; (2) large-scale datasets; (3) high-dimensional datasets; (4) datasets with...

AI Technical Summary

Problems solved by technology

Through the above analysis, the current grid clustering algorithm has the following problems in grid processing: (1) Clusters with complex boundary shapes cannot be effectively identified; (2) The complexity of th

Images