Clustering method and device

Abstract

The invention provides a clustering method, and aims at carrying out category classification on points of a certain set. The method comprises the following steps of: obtaining appointed result attributes of partial points in the set, wherein the appointed result attributes are used for limiting category classification results of the points; and operating a clustering algorithm by taking the appointed result attributes of the partial points in the set as constraint conditions, so as to obtain clustering results according with the appointed result attributes. Through the clustering method and device, analyzers can influence clustering results through setting appointed result attributes of points, so that the clustering results are guided to directions according with analysis demands, and theconvergence speed of clustering analysis in improved while the clustering quality is enhanced.

Description

technical field [0001] The present application relates to the technical field of data processing, in particular to a clustering method and device. Background technique [0002] With the increasingly widespread application of information technology, data is continuously generated in any link of production and life, and the amount of collected and stored data is constantly expanding. Finding meaningful patterns or rules from massive data through data mining, providing auxiliary support for information management, query optimization, business decision-making and process control, has become an important way to use data. [0003] Cluster analysis is a means of data mining. Clustering is to divide the data set (also called point set, each point is a data record) into multiple categories according to the inherent similarity of data records, so that the points in the category are similar The higher the degree, the smaller the point similarity between categories. At present, there ...

AI Technical Summary

Problems solved by technology

In this way, it often takes many attempts to obtain a satisfactory clustering result, and sometimes the desired classification result cannot be obtained even after multiple attempts, and the efficiency and quality of clustering analysis cannot be guaranteed.

Images