Filtering of anomalous weather events over the region of Tamil Nadu

• Published in: 2017 IEEE International Conference on Intelligent Techniques in Control, Optimization and Signal Processing (INCOS) pp. 1 - 7
• Main Authors: Piruthevi, C., Selvi, C.S. Kanimozhi
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-03-2017
• Subjects: Anomalous weather events; Anomaly Frequency Method; Association rule mining; Conferences; Cyclones; Data mining; Data summarization; Humidity; Prediction algorithms; Signal processing algorithms
• DOI: 10.1109/ITCOSP.2017.8303117

An anomaly is a short yet significant deviation from the normal behaviour. But the main problem identified in the domain Spatio-Temporal Data Mining is the early prediction of anomalous weather events as it has a great impact on the economy and life of people. People from different parts of the world suffer from the severe consequences of anomalous weather events like heavy rainfall, drought and snowfall. Here, Anomaly Frequency Method (AFM) is used for extracting the anomalous weather events over a particular region. AFM method identifies the extreme deviation in the weather parameters. From the obtained anomalous weather events, association rule mining is made using Apriori, Predictive Apriori and Generalized Sequential Pattern algorithm for the generation of different weather patterns. Data summarization about the extreme weather events is made which results in providing the information about their occurrences based on the deviations found in the parameters used.