文献类型：期刊文献

中文题名：Predicting Precipitation Events Using Gaussian Mixture Model

英文题名：Predicting Precipitation Events Using Gaussian Mixture Model

作者：Haitian Ling[1];Kunping Zhu[1]

机构：[1]School of Science, East China University of Science and Technology, Shanghai, China;[2]Department of Statistics, University of Illinois at Urbana-Champaign, Champaign, IL, USA

年份：2017

卷号：5

期号：4

起止页码：131

中文期刊名：Journal of Data Analysis and Information Processing

外文期刊名：数据分析和信息处理（英文）

语种：英文

中文关键词：Gaussian;Mixture;Model;Classification;EM;Algorithm;Precipitation;Event

外文关键词：Gaussian Mixture Model;Classification;EM Algorithm;Precipitation Event

摘要：In this paper, a Gaussian mixture model (GMM) based classifier is described to tell whether precipitation events will happen on a certain day at a certain time from historical meteorological data. The classifier deals with a two-class classification problem where one class represents precipitation events and the other represents non-precipitation events. The concept of ambiguity is introduced to represent cases where weather conditions between the two classes like drizzles, intermittent or overcast are more likely to happen. Six groups of experiments are carried out to evaluate the performance of the classifier using different configurations based on the observation data released by Shanghai Baoshan weather station. Specifically, a typical classification performance of about 75% accuracy, 30% precision and 80% recall is achieved for prediction tasks with a time span of 12 hours.
In this paper, a Gaussian mixture model (GMM) based classifier is described to tell whether precipitation events will happen on a certain day at a certain time from historical meteorological data. The classifier deals with a two-class classification problem where one class represents precipitation events and the other represents non-precipitation events. The concept of ambiguity is introduced to represent cases where weather conditions between the two classes like drizzles, intermittent or overcast are more likely to happen. Six groups of experiments are carried out to evaluate the performance of the classifier using different configurations based on the observation data released by Shanghai Baoshan weather station. Specifically, a typical classification performance of about 75% accuracy, 30% precision and 80% recall is achieved for prediction tasks with a time span of 12 hours.