JOURNAL OF INFORMATION SYSTEM RESEARCH (JOSH)
Vol 5 No 1 (2023): Oktober 2023

Klasifikasi Sentimen Masyarakat Di Twitter Terhadap Prabowo Subianto Sebagai Bakal Calon Presiden 2024 Menggunakan M-KNN

Abdul Halim (Universitas Islam Negeri Sultan Syarif Kasim Riau, Pekanbaru)
Yusra Yusra (Universitas Islam Negeri Sultan Syarif Kasim Riau, Pekanbaru)
Muhammad Fikry (Universitas Islam Negeri Sultan Syarif Kasim Riau, Pekanbaru)
Muhammad Irsyad (Universitas Islam Negeri Sultan Syarif Kasim Riau, Pekanbaru)
Elvia Budianita (Universitas Islam Negeri Sultan Syarif Kasim Riau, Pekanbaru)



Article Info

Publish Date
26 Oct 2023

Abstract

Presidential elections are held every five years and each presidential candidate will get support from several political parties to run for candidacy in the election. In a multi-party system, the number of parties participating in the election is very large, so that the perspectives of voters on political actors, including presidential candidates who will advance in the 2024 elections, are varied. The survey results from Polling Indonesia (SPIN) conducted from 7 to 16 October 2022 show that Prabowo Subianto has the highest electability with a score of 31.6%, based on a national leadership survey. In this study, a test was carried out by classifying tweet data from the public collected on the Twitter application from January to December 2022 using the Modified k-Nearest Neighbor method to analyze public sentiment regarding the upcoming election. Data collected as many as 2,100 data with positive and negative categories related to "Presidential Candidate" and "Prabowo Subianto" and the implementation of the Modified k-Nearest Neighbor classification was carried out using Google Colab. Based on the results of the confusion matrix test from the Modified k-Nearest Neighbor classification with three comparisons made (ie comparisons 70%:30%, 80%:20% dan 90%:10%) and using K=3, 5, 7, 9, 11 when testing a comparison of 90:10 at K=3 the highest accuracy results were obtained with a value of 93,3%.

Copyrights © 2023






Journal Info

Abbrev

josh

Publisher

Subject

Computer Science & IT Decision Sciences, Operations Research & Management

Description

Artikel yang dimuat melalui proses Blind Review oleh Jurnal JOSH, dengan mempertimbangkan antara lain: terpenuhinya persyaratan baku publikasi jurnal, metodologi riset yang digunakan, dan signifikansi kontribusi hasil riset terhadap pengembangan keilmuan bidang teknologi dan informasi. Fokus Journal ...