Map reduce technique for parallel-automata analysis of large scale rainfall data

Vast scale rainfall information assumes an imperative part in farming field thus early expectation of rainfall is important for the better finan-cial development of a nation. Rainfall expectation is an expert among the most troublesome issue far and wide in a year back. This data is generally secured in the unstructured course of action. Along these, tremendous measure of data has been accumulated and archived. Thus, storage and handling of such tremendous information for accurate rainfall forecast are a major test. Big Data innovation like Hadoop have developed to fathom the difficulties and issues of huge information utilizing distributed computing. Till date few examinations have been accounted for on the preparing of vast scale rainfall information utilizing MapReduce. In this paper, the huge scale rainfall information is anticipated by utilizing MapReduce system which plays out the capacities which are required and diminishes the task to get proficient ar-rangements through taking the information and isolating into smaller tasks. At that point, the three Regression Automata (RA) algorithms such as Linear Regression automata, Support Vector Regression Automata and Logistic Regression Automata. are utilized to forecast the future esteem of large scale rainfall data. The proposed framework serves as a tool that takes in the rainfall information from diminished information as input and predicts the future rainfall. The outcomes obviously demonstrate that the all the three RA models can anticipate the rainfall productively in different terms, such as, error rate, coefficients and mean square error.