Weather Analysis And Prediction Based On Data Mining

With the rapid development of information technology in our country,meteorological departments have accumulated a large amount of data,and behind the surge of meteorological data hide a large number of important information which is little known.How to find useful information from these surge and large amount of data has become a hot issue in data mining and meteorological expert research nowadays.On the other hand,it is well known that meteorology is a hot topic.Information is highly related to people's lives,seriously affecting people's production and lifestyle,and closely related to people's vital interests.Based on the above research background,starting from the temperature that people are most concerned about in daily life,this paper establishes a model by studying the relationship among several factors affecting the temperature change: rainfall,air pressure,water pressure and so on,uses the model to predict,and studies the effect of the model by studying the error between the predicted value and the real value.This paper focuses on two methods to study this topic.The following two methods are briefly introduced:Firstly,the decision tree method is used to evaluate the temperature data,which mainly considers six indices which are closely related to the temperature change,such as rainfall and monthly average temperature.The decision tree model of monthly average temperature and the decision tree model divided by season are established.By comparing the two methods,we find that the prediction effect of the decision tree model based on monthly average temperature is important as a whole.It is superior to the decision tree model based on season,but the decision tree model based on season is superior to the decision tree model based on monthly average temperature in predicting summer temperature.Secondly,this paper fully considers the method of dealing with this problem according to the time factor,uses the simple seasonal model and ARIMA model commonly used in time series to analyze,model and forecast the data involved in this paper,and finally finds that the prediction effect of ARIMA model is better than that of simple seasonal model.Finally,the author summarizes the role of this model in the case analysis,and also analyses some problems and shortcomings in the process.At the same time,these problems also point out the direction for future research.