Clustering method for predicting campaign results based on voter and candidate characteristics

Jonson Manurung (STMIK Pelita Nusantara, Medan, Indonesia)
Sethu Ramen (STMIK Pelita Nusantara, Medan, Indonesia)
Logaraj Logaraj (STMIK Pelita Nusantara, Medan, Indonesia)

Jurnal Mantik
Vol. 7 No. 2 (2023): Agustus: Manajemen, Teknologi Informatika dan Komunikasi (Mantik)

This research applies clustering method with K-Means algorithm to analyze voter preferences and predict campaign outcomes based on voter and candidate characteristics in the context of political elections. By collecting and processing data on age, education, occupation, and candidate preferences, we apply K-Means to cluster voters into groups with similar patterns. The cluster results reveal similar political views and candidate preferences within each group of voters. By correlating the cluster results with previous election data, we are able to predict campaign outcomes with an accuracy that is beneficial for more careful and effective campaign strategies. This research contributes to a deeper understanding of the use of clustering methods in the context of political elections and its relevance in formulating successful campaign strategies.