论文部分内容阅读
随着降雨量预测在中国的气象预报行业中日趋重要,降雨量预测的方法也越来越多。由于云平台可以有效地提高预测的效率和准确率,云平台也逐渐被应用到气象行业。目前我们运用的降雨量预测方法要求属性之间独立,但是很多气象要素之间并不独立,这就降低了预测的准确性。因此,结合并利用模糊集理论的相关知识,提出了一个基于云平台的半朴素贝叶斯预测降雨量的方法。为证明预测的准确性和高效性,建立了一个预测模型,用气象站提供的气象数据预测下个月的降雨量。实验结果证明,建立的模型与先前的模型相比,具有更高的预测准确性和效率。
As rainfall forecasting becomes more important in the meteorological forecasting industry in China, there are more and more ways to forecast rainfall. As the cloud platform can effectively improve the prediction efficiency and accuracy, cloud platform is gradually being applied to the weather industry. At present, the rainfall forecast method we use requires independence of attributes, but many meteorological elements are not independent, which reduces the accuracy of forecasting. Therefore, based on the related knowledge of fuzzy set theory, a method of semi-naive Bayesian prediction of rainfall based on cloud platform is proposed. In order to prove the accuracy and efficiency of the prediction, a prediction model was established to predict the rainfall of next month using the meteorological data provided by the weather station. The experimental results show that the established model has higher prediction accuracy and efficiency than the previous model.