Clustering Analysis of Extreme Temperature Based on K-means Algorithm

Abstract

The research is based on the 2 GB extreme climates data recorded from 825 meteorological stations in mainland cities of China which was published online by the National Climate Center of the China Meteorological Administration. It analyzes the changes of extreme temperature in different time in the country and the spatial distribution of extreme temperature. The k_means clustering algorithm in data mining is used to study the regional aggregation of extreme temperature time across the country, and the experimental results are visually displayed and introduced. The analysis results show that the frequency and area of extreme high temperature events are increasing over time throughout the country. Extreme temperature events not only have regional aggregation, but also the occurrence of extreme temperature events with regional mobility. In this paper, the Apriori Algorithm of mining association rules in data mining technology is used to study the regional association pattern of extreme climate events in small areas. On this basis, the K-means Clustering Algorithm is used to identify and express the regional aggregation of time and space data in the extreme temperature events nationwide, which provides a new idea and method for the study of extreme climate events.