Perbandingan Kinerja Naive Bayes dan KNN Dalam Klasifikasi Sentimen Ulasan Film Horor

Cantriya Simbolon; Maria Angelina Lubis; Sardo Pardingotan  Sipayung

doi:10.31316/jdi.v15i1.464

Perbandingan Kinerja Naive Bayes dan KNN dalam Klasifikasi Sentimen Ulasan Film Horor

Authors

Cantriya Simbolon Universitas Katolik Santo Thomas
Maria Angelina Lubis Universitas Katolik Santo Thomas
Sardo Pardingotan Sipayung Universitas Katolik Santo Thomas

DOI:

https://doi.org/10.31316/jdi.v15i1.464

Keywords:

Analisis Sentimen, Naive Bayes, K-Nearest Neighbors, Rapid Minder, Film Horor

Abstract

Lonjakan ulasan film horor di platform digital memerlukan sistem klasifikasi otomatis untuk
memahami sentimen penonton secara efisien. Penelitian ini bertujuan membandingkan kinerja
algoritma Naive Bayes dan K-Nearest Neighbors (KNN) dalam mengklasifikasikan sentimen ulasan
film horor berbahasa Inggris. Metodologi penelitian melibatkan pengolahan 3.000 data dari Kaggle
menggunakan perangkat lunak RapidMiner, dengan tahapan pra-pemrosesan meliputi pembobotan
TF-IDF, tokenization, filtering, dan stemming. Pengujian dilakukan melalui skema 10-fold cross
validation untuk menjamin stabilitas hasil. Temuan penelitian menunjukkan perbedaan performa
yang signifikan, di mana Naive Bayes meraih akurasi sebesar 88,53%, jauh mengungguli KNN yang
hanya mencapai 40,47%. Rendahnya akurasi KNN disebabkan oleh kompleksitas perhitungan jarak
pada data teks berdimensi tinggi. Disimpulkan bahwa Naive Bayes merupakan model yang lebih
reliabel dan efektif untuk klasifikasi sentimen ulasan film horor. Hasil ini memberikan kontribusi
berupa rekomendasi algoritma optimal bagi pengembangan sistem analisis opini otomatis.

Downloads

PDF (Bahasa Indonesia)

Published

2026-03-01 — Updated on 2026-03-09

Versions

Issue

Vol. 15 No. 1 (2026): Vol. 15 No. 1 (2026)

Section

Articles

License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Authors who publish with this journal agree to the following terms:

Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-ShareAlike 4.0 International License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).