Research on Association Rules Data Mining Based on Improved K-means Algorithm

Abstract

The process of data mining is essentially the process of finding value in huge and random data. With the development of the times, information technology has continuously entered various fields, which has greatly expanded the amount and types of data that can be used for mining. In data mining, association rules are the association relationships between data. By grasping the association rules between massive data, users can have a special understanding of the data to assist judgment. This paper improves the K-means algorithm based on the three selected test indicators, eliminates redundant rules through certain criteria, obtains the initial point through triangle iteration under the K-means rule, and clusters the rules. Through the research in this paper, it can be confirmed that the aggregation of similar rules can help data analysts quickly obtain useful rules and help the further advancement of the analysis process.