A Model for Empirical Earthquake Prediction and Analysis in a Data Intensive Environment
Faraz Ahmad1, Om Ashish Mishra2, Shivam Bhagwani3, Jabanjalin Hilda J4
1Faraz Ahmad, Department of Computer Science Engineering, Vellore Institute of Technology, Vellore, India.
2Om Ashish Mishra, Department of Computer Science Engineering, Vellore Institute of Technology, Vellore, India.
3Shivam Bhagwani, Department of Computer Science Engineering, Vellore Institute of Technology, Vellore, India.
4Jabanjalin Hilda J, Vellore Institute of Technology, Vellore, India.
Manuscript received on 05 April 2019 | Revised Manuscript received on 14 April 2019 | Manuscript Published on 24 May 2019 | PP: 282-288 | Volume-8 Issue-6S3 April 2019 | Retrieval Number: F10580486S319/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Characteristic perils like earthquakes are for the most part the consequence of spreading seismic waves underneath the surface of the earth. Tremors are dangerous absolutely in light of the fact that they’re erratic, striking without warning, triggering fires and tsunamis and leading to deaths of countless individuals. If researchers could caution people in weeks or months ahead of time about seismic disturbances, clearing and different arrangements could be made to spare incalculable lives. An early identification and future earthquake prediction can be achieved using machine learning models. Seismic stations continuously gather data without the necessity of the occurrence of an event. The gathered data can be used to distinguish earthquake and non-earthquake prone regions. Machine learning methods can be used for analyzing continuous time series data in order to detect earthquakes effectively. The pre-existing linear models applied to earthquake problems have failed to achieve significant amount of efficiency and generate overheads with respect to pre-processing. The proposed work exploits parallel processing in Hadoop by using the various frameworks like Pig-Hive optimization, Map Reduce and Impala, in order to mine and analyze earthquake data to propose a model for predicting future earthquakes.
Keywords: Earthquake, Pig-hive, Prediction, Classification, Machine Learning, Impala.
Scope of the Article: Classification