Analysis of Consumer Satisfaction Levels with GoRide Services Using the Support Vector Machine (SVM) Classification Method

The online transportation service that is popular and often used by the public is Gojek. Every customer has different feedback on Gojek’s services. The quality of services provided can influence the level of customer satisfaction. Therefore, it is necessary to analyze the prediction of the level of satisfaction of GoRide. This study uses the Support Vector Machine (SVM) method to predict it. The object was the students of IST AKPRIND Yogyakarta. The results of the classification of the level of consumer satisfaction on GoRide use the SVM method, which uses a linear kernel function with Cost = 0.1 and testing data of 25%, producing an accuracy rate of 100%. SVM with RBF kernel function, with a value of Cost = 100 and gamma = 0.00026, makes an accuracy rate of 88.46%. The classification results are that 76.69% of respondents stated that they were satisfied and delighted with the GoRide service. Meanwhile, a total of 23.30% of respondents indicated that they were dissatisfied and very dissatisfied with GoRide’s services. This proves that the SVM method can predict the classification of the level of consumer satisfaction with GoRide services very well.