Predictive Analytic Klasifikasi Penentuan Tarif Sewa Bus

Abstract

Currently, the bus rental business has become the choice of consumers in traveling, because of the decision of flexibility and better availability. The government regulates that consumer and business owner agreements determine bus rental rates without routes. In this study intends to do clustering from the history of raw data that already exists before. Data is obtained from companies in the form of spreadsheet files originating from non-information systems. The raw data is combined and normalized, to eliminate the noise data and the data is not abnormal. The clustering results using the K-Means algorithm and Louvain clustering produce several tariff groups that can be used as a reference for determining fare. In this paper also concludes about unbalanced data, which can cause data clustering errors.