Classification Using C4.5 Algorithm in Election Participation Prediction

Abstract

The General Election of the President, DPD, DPR, Provincial DPRD and Regency / City DPRD in 2019, in Indonesia, is carried out by an independent institution namely the General Election Commission (KPU), where there is a measure of the success of the holding by holding the principle of direct, general, free of secrecy. Another component of the election will be the implementation of contestants and voters. In the electoral factor, this is also a measure for success in the overall implementation process, which is a success if the participation of the community is high in the holding of elections. And conversely, if the community participation is low, one of them is the level of public trust in the organizer (government) decreases. Analysis of classification mining data with C4.5 algorithm places the prediction class “Present” and “Not Present” on the permanent voter list (DPT) taken from form C7 (voter attendance list form). C7 data was obtained from polling stations (TPS) located in the village of Wonokasian, Wonoayu District, Sidoarjo Regency, East Java. The number of datasets is 4249 instances, and the number of attributes is 10. With split datasets, for training data are 2965 instances, and testing data are 1284 instances. From testing, the testing data obtained the results of true or appropriate accuracy of 1151 instances (90%) and obtained incorrect or incompatible data as many as 134 instances (10%). This shows a good value for the results of 90% of the corresponding data in the prediction class in the Presidential Election, DPD, DPR, Provincial DPRD, and Regency / City DPRD in 2019.