Using of Meteorological Data to Estimate the Multilevel Clustering for Rainfall Forecasting

Abstract

The world's agriculture, especially third world agriculture, depends upon the seasonal rainfall pattern. Recent erratic changes in rainfall pattern lead toward low agriculture production, thus creating food insecurity for an ever-increasing world population. Flood, drought, and famine are the consequences of these changing patterns. Weather patterns and forecasting are the main topics of atmospheric science. In meteorological research, statistical and numerical analysis are crucial. By utilising forecasting models and weather forecasting tools, meteorological data will be used to predict changes in climatic patterns.  Data mining techniques have more scope to discover future weather patterns by analyzing past weather dimensions. In our study two techniques Multiple Linear Regression (MLR) and Expectation Maximization (EM) clustering algorithms are combined for rainfall forecasting. MLR interprets most important parameters of rainfall for clustering algorithm. When applied to selected partitioned attributes, the EM clustering algorithm will detect correctly and incorrectly clustered instances. By analysing previous meteorological observations, the model was able to forecast less rainfall, medium rainfall, and heavy rainfall. Standard deviation is used as a measure of error correction to improve the  cluster results. Data normalization helps to improve model performance. These findings are useful to determine future climate expectation. Partitioned  procedure will be applied on selected attributes and EM clustering algorithm was then executed on partitioned data set to determine less, medium and high rainfall.