Classification tide levels in Semarang City use support vector machine

• Published in: Journal of physics. Conference series Vol. 1217; no. 1; pp. 12103 - 12108
• Main Authors: Sugito, Safitri, D, Mustafid, Ispriyanti, D, Prahutama, A
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 01.05.2019
• Subjects: Accuracy; Classification; Kernel functions; Physics; Polynomials; Rainfall; Support vector machines; Tidal flooding; Training; Wind speed
• ISSN: 1742-6588; 1742-6596
• DOI: 10.1088/1742-6596/1217/1/012103

As the capital of Central Java, Semarang is a coastal town located on the north coast of Java. This geographical feature exposes the city to environmental problems such as tidal flood (locally known as rob). Therefore, it is high at stake for the city to classify a certain model to determine the high of tide. One of the expert classification methods of machine-based learning is support vector machine (SVM), it is classified as the nonparametric machine which does not require any assumptions. The classification using SVM requires a kernel as a weight to determine support vector data to classify. Therefore, this study uses Kernel of polynomial and radial base function. As for variables of tidal classification, this study used wind speed and rainfall. On the basis of the analysis, the maximum tide level classification accuracy was carried out on the distribution of 80: 20 training and testing data resulting in a classification accuracy of 69.42%. Classification accuracy was determined by the distribution of training and testing data.