Performance of Classification Algorithms with Weka and Spark Tools
Kasarapu Ramani · Journal of Emerging Technologies and Innovative Research · 2018
Because of the extensive number of impacting factors, it is hard to foresee the natural disasters such as earthquake. Analysts are working seriously on earthquake forecast. Death toll and property can be limited with earthquake prediction. In this paper, the performance of the methods such as the Decision Tree and Naive Bayes has been compared to find the algorithm that best fits the earthquake prediction. Also the performance of these classification algorithms is tested on two different tools such as Weka and Spark. The performance is expressed in terms of parameters correctly classified instances, incorrectly classified instances, errorrate and precision. The Decision Tree algorithm has given improved precision than the Naive Bayes by approximately 3-4 percentage. It can be concluded by the analysis that the performance of both the algorithms was high when performed in the Spark tool as compared to the Weka tool.