K-Means Cluster Using Rainfall and Storm Prediction in Machine Learning Technique

Abstract

Listen

Translate article

Data Mining involves extracting meaningful information from the available data in a user understandable manner. Its role is to analyze voluminous data that is being often assembled. Using the approach of Data mining techniques various business related queries can be attended which formerly were extremely time-consuming to answer. There exist uncontrollable natural disasters that critically hampers and costs human life, environment and revenue material. Natural calamities like heavy rainfall and floods cannot be well predicted until it happens, also it’s beyond one’s power to control them. The aftereffect or destruction caused by these calamities prevails for many years. The term disaster is a result of a vulnerable condition caused by heavy rainfall, flood or storm that can have intense effect at a smaller scale such as a village or at a larger scale such as city or state. Clustering model that was developed before confronted the issue of time complexity, low processing speed and were inappropriate for huge datasets. The current research work proposes the approach of K means clustering that is a subset of ML (machine learning) techniques that are capable to process huge datasets and performs quick computation compared to rest of the clustering model. Various stages in this proposed system include Dataset Collection, Pre-processing, Feature selection, K-means clustering. Among these, the K-means clustering tool which is actually a subset of data-mining and ML approach is employed to cluster observations in the form of groups. It’s a form of unsupervised learning that rectifies the clustering problem. The results reveal that the K-means clustering tool performs clustering faster than the other existing technique.