Proposing A New Methodology For Weather Forecasting By Using Big Data Analytics

Abstract

Big data has described an enormous quantity of data which needs new technologies to make potential to obtain value from it by analysis and capturing method. Data Analytics often includes scrutinizing past traditional data to research potential trends. Weather prognostication has been one of the most fascinating and exciting domain, and it performs an essential role in aerography. The weather situation is the state of the atmosphere at a given time regarding weather variables like wind direction, rainfall, cloud conditions, pressure, temperature, thunderstorm, etc. The Big data obtained by NCDC (National Climatic Data Center) has received over more than 116 weather locations and more than 1000 observations centers. The data produced by them is unstructured which grows a challenging job to explain it. In this paper, these enormous amounts of data have loaded onto the Apache Pig, Hadoop Distributed File System, Apache Hive is to process the data, which utilizes mappers and reducers to process the data. The above dataset has explained by using given methods and the final output of this project in the form of maximum, minimum and average temperature according to the given time and date.