Research Article
Research and Application of Summer High Temperature Prediction Model Based on CART Algorithm
@INPROCEEDINGS{10.1007/978-3-319-72823-0_30, author={Yujie Guan and Wei Wang and Fengchang Xue and Shoudong Liu}, title={Research and Application of Summer High Temperature Prediction Model Based on CART Algorithm}, proceedings={5G for Future Wireless Networks. First International Conference, 5GWN 2017, Beijing, China, April 21-23, 2017, Proceedings}, proceedings_a={5GWN}, year={2018}, month={1}, keywords={CART High temperature effective accumulated temperature Summer high temperature forecast}, doi={10.1007/978-3-319-72823-0_30} }- Yujie Guan
Wei Wang
Fengchang Xue
Shoudong Liu
Year: 2018
Research and Application of Summer High Temperature Prediction Model Based on CART Algorithm
5GWN
Springer
DOI: 10.1007/978-3-319-72823-0_30
Abstract
In this paper, the average summer high temperature effective accumulated for many years is used as a judge of the extent of the hot summer temperatures of standards. Based on data mining, the CART algorithm is applied to analyze the relationship between high temperature and some climatic factors such as the East Asian summer monsoon index, summer India Burma trough, the summer North Atlantic Oscillation (NAO), Equatorial Pacific sea surface temperature and so on. The high-temperature forecasting model is established with the setup of the high temperature prediction rules. The data of summer maximum temperature in summer in Zhangzhou, Fujian Province from 1955 to 2012 are selected to calculate the summer hot temperature of 58a. Then, multiple climatic factor data of the same period is given to the input variable, and 46 years of data is randomly selected to get 10 classifications of rule sets, resulting in the achievement of the accuracy rate to 91.49%. With the remaining data of 12a test, the accuracy rate reaches 91.67%. In general, the results of this paper validate the feasibility and validity of the high temperature prediction model, and provide a new idea for the study of the catastrophic weather model.