PREVENTIVE DETECTION SYSTEM PADA LAPORAN FRAUD MENGGUNAKAN MODEL TIMESERIES PADA KANTOR BEA CUKAI

Abstract

State revenue from the excise sector has increased every year because cigarette excise is the biggest contributor to state revenue. Fraud or cheating occurs because of motivation and encouragement from various parties. The techniques of fraud are varied, ranging from outsmarting principles to committing illegal acts were then led to the bankruptcy of the company. According to data from the Directorate General of Customs and Excise, which has its head office in Jalan Jenderal A Yani Rawamangun, East Jakarta, fraud was committed by tobacco senders who did not have excise stamps. The problem in this study is how is the preventive detection system on fraud reports at the customs office. The research objective is to obtain preventive detection measures for fraud reports. The results of time series modeling have a score of 1,006 for the training data and a score of 1 for the data testing showing that the prediction results have a high degree of accuracy.