Water Quality Monitoring using Modified Support Vector Machine Algorithm in Comparison of Accuracy with Artificial Neural Network
S. Vaishnavi, G. Ramkumar · 2023
This study focuses on the prediction of water quality monitoring based on the parameters collected from the water data set by using the Innovative Support Vector Machine (SVM) method in order to evaluate the prediction accuracy with an Artificial Neural Network (ANN). This analysis required the collection of 40 samples, split evenly between two sets of 20 samples each. Support Vector Machine (SVM) is what Group 1 utilizes, whereas Group 2 makes use of an artificial neural network (ANN). The procedure for this study has resulted in the data set being imported and the code for the Support Vector Machine being developed. This was accomplished with the assistance of the anaconda program, and the jupyter notebook has been started. The sample size is derived using the numbers received from the earlier research with the assistance of an online statistical analysis tool using the pretest power of 80% and the alpha value of 0.05 as the values to be used in the calculation. The results of the simulation showed that the Support Vector Machine (SVM) Algorithm had an accuracy of prediction of 90.4%, while the Artificial Neural Network (ANN) had an accuracy of prediction of 72.1% with a significance of 0.020, which is less than 0.05. When it comes to discovering the prediction accuracy for water quality monitoring, the Support Vector Machine (SVM) performs noticeably better than the Artificial Neural Network (ANN) when using the dataset that is provided.