Water quality monitoring using modified support vector machine algorithm in comparison of accuracy with Adaboost classifier

The primary objective of this study is to compare the Adaboost method with the Conventional SVM technique in order to provide an accurate prediction about whether or not the fluid is fit for usage by humans. Two sets of twenty samples each were used for this study, totaling forty data. For group 1, we have SVM, while for group 2, we have Adaboost. With a prior test efficiency of 80% and an alpha value of 0.05, the sampling is computed using the results of current research as well as the system clincalc.com. According to the workflow to implement the support vector machine (SVM), the code is implemented and compiled in the Anaconda software with the Jupyter Notebook installed and launched successfully. The proposed work support vector machine (SVM) has got a higher accuracy of 90.75% whereas the existing Adaboost classifier has got an accuracy of 75.14%. The significance 0.000 obtained is less than 0.05 value. Therefore, the suggested study concludes that, when comparing the precision of water quality, SVM outperforms the current classifier, the Adaboost classifier.