@article{ref1,
title="Prediction of extreme weather events using machine learning technique",
journal="International journal of applied engineering research",
year="2019",
author="KanimozhiSelvi, Dr C. S. and Sowmiya, G.",
volume="14",
number="4",
pages="924-929",
abstract="In climate applications, an event is an instance in time when a significantand persistent change occurs. Spatiotemporal Data Mining is used in the department of meteorology for the prediction of all kinds of weather data and global changes. The main problem is prediction of extreme weather events over a particular region. This system will predict theweather based on parameters such as air temperature, relative humidity, specific humidity, U-wind, Vwind and omega.   Here Anomaly frequency method (AFM) technique is used for extracting the anomalous weather events over a particular region. The AFM is an efficient technique in extracting the features which discriminate extreme events and non-extreme events. DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is used to find the co-variation among different climate extremes and their relationship to other climate phenomena. DBSCAN and K-Means clustering algorithmsare performed on weather parameters and the extreme and non-extreme weather events are visualized.   The analysis techniques like homogeneity, completeness, V- measure, Adjusted Rand Index (ARI), Adjusted Mutual Information (AMI) are used to calculate the accuracy for all the parameters. Also the results are validated with the realtime weather events like BOB, Thane and Vardah.     Keywords: Extreme weatherevents; Anomaly frequency method; Density-Based Spatial Clustering of Application with Noise<p /> <p>Language: en</p>",
language="en",
issn="0973-4562",
doi="",
url="http://dx.doi.org/"
}