A Hierarchical Clustering Method for Tourist Destination Data

Abstract

A hierarchical clustering method for tourist destination data based on regional growth and competition in a variable-scale data density space is different from the previous method. The present invention adopts the idea of ​​hierarchical clustering and divides the clustering process into three levels. The first-level clustering uses a distance threshold R1 based on Euclidean distance to divide objects into a certain number of sub-classes to simplify the algorithm and reduce the complexity. Then the second stage uses the method of spatial data region growth, using the obtained cluster centers as growth seeds, and grows under the growth criterion until the stopping condition is reached to solve the problem of variable-scale data density clustering. The last third level is based on the idea of ​​competition and the principle of density similarity, calculates the weight between the cluster centers, and adopts appropriate rules to merge the clusters to solve the problem of non-convex data clustering. Compared with other clustering algorithms, the method of the invention can greatly improve the accuracy of clustering on the basis of reducing the complexity, has obvious advantages in processing massive data, and can better meet the needs of practical engineering applications.

Description

technical field [0001] The invention relates to the field of hierarchical clustering, and specifically uses a method based on region growth and competition to improve the clustering method of variable-scale density data. Background technique [0002] Data mining is a hot issue in the field of artificial intelligence and database research. Cluster analysis is an important branch of data mining. As a tool for data analysis, clustering has been widely used in various fields. Clustering is the process of dividing a physical or abstract collection into classes consisting of similar objects. Clustering originated from taxonomy, but it is different from classification. The difference between clustering and classification is that the classes required for clustering are unknown and unsupervised. Clustering algorithms are roughly divided into (1) partition-based methods, such as K-means algorithm, K-medoids algorithm, etc.; (2) hierarchical-based methods, such as BIRCH algorithm, CU...

AI Technical Summary

Problems solved by technology

Aiming at the defect that multi-scale data is mostly limited to space, the present invention uses multi-level analysis to propose a new distance-based multi-level clustering algorithm according to actual needs, and uses multi-level fast non-convex clustering to solve multi-scale density data. clustering problem

Images