Big Data Technologies using SVM (Case Study: Surface Water Classification on Regional Water Utility Company in Surabaya)

Abstract

How important to the role of water for the survival of living beings, not only for human but also the other living beings need water as one of the elements that support the continuity of life in every living creature. To maintain the necessity of water resources such as river, recently the need for monitoring systems that able to take the parameter of water quality using sensors important. In the previous paper, we built the Internet of Things to get the data using a passive sensor and an active sensor. As additionally, we built Big-Data system equipped with machine learning algorithm that can perform water quality classification with the Support Vector Machine method. This system monitoring every activity in the Karang Pilang area and applying classification. The result of this system that the big data system can perform the classification of river water quality in interactive and accurate. The result discusses that we were able to classify by using Support Vector Machine with accuracy level 0.9138 by using Linear kernel and 0.8372 by using RBF kernel. From the ROC result, we achieved AUC value until 0.93. It’s mean we achieved an excellent result.