ANALISIS PERBANDINGAN KINERJA ALGORITMA SUPPORT VECTOR MACHINE DAN NAÏVE BAYES TERHADAP SENTIMEN PUBLIK PADA PROGRAM MAKAN BERGIZI GRATIS DI MEDIA SOSIAL X

NISA ADELIA, NPM 2209100090 (2026) ANALISIS PERBANDINGAN KINERJA ALGORITMA SUPPORT VECTOR MACHINE DAN NAÏVE BAYES TERHADAP SENTIMEN PUBLIK PADA PROGRAM MAKAN BERGIZI GRATIS DI MEDIA SOSIAL X. Skripsi thesis, Universitas Labuhanbatu.

	Text COVER.pdf Download (829kB)
	Text BAB I.pdf Download (259kB)
	Text BAB II.pdf Download (745kB)
	Text BAB III.pdf Download (4MB)
	Text BAB IV.pdf Download (991kB)
	Text BAB V.pdf Download (178kB)
	Text DAFTAR PUSTAKA.pdf Download (230kB)

Abstract

Penelitian ini bertujuan untuk membandingkan kinerja algoritma Support Vector Machine (SVM) dan Naïve Bayes (NB) dalam mengklasifikasikan sentimen publik terhadap Program Makan Bergizi Gratis di media sosial X. Data dikumpulkan melalui X API dan dilabeli ke dalam tiga kategori sentimen, yaitu positif, negatif, dan netral. Tahapan prapemrosesan meliputi case folding, tokenizing, stopword removal, dan stemming, kemudian direpresentasikan menggunakan metode Term Frequency–Inverse Document Frequency (TF–IDF). Model dikembangkan menggunakan SVM dan NB serta dievaluasi menggunakan stratified 10-fold cross validation dengan metrik accuracy, precision, recall, dan F1 score. Hasil penelitian menunjukkan bahwa SVM memperoleh tingkat akurasi sebesar 0,79, sedangkan Naïve Bayes memperoleh akurasi sebesar 0,68. Perbedaan ini menunjukkan bahwa SVM memiliki kemampuan klasifikasi yang lebih baik dalam menangani data teks berdimensi tinggi dan distribusi kelas yang tidak seimbang. Penelitian ini menyimpulkan bahwa SVM lebih efektif digunakan dalam analisis sentimen kebijakan publik berbasis media sosial, sementara Naïve Bayes tetap memiliki keunggulan dalam efisiensi komputasi dan kesederhanaan model. Kata kunci: Analisis Sentimen, Support Vector Machine, Naïve Bayes, TF–IDF, Media Sosial X =================================== This study aims to compare the performance of the Support Vector Machine (SVM) and Naïve Bayes (NB) algorithms in classifying public sentiment toward the Free Nutritious Meal Program on social media platform X. The data were collected using the X API and labeled into three sentiment categories: positive, negative, and neutral. The preprocessing stages include case folding, tokenizing, stopword removal, and stemming, followed by feature representation using the Term Frequency–Inverse Document Frequency (TF–IDF) method. The models were developed using SVM and NB and evaluated using stratified 10-fold cross validation with performance metrics including accuracy, precision, recall, and F1 score. The results show that SVM achieved an accuracy of 0.79, while Naïve Bayes achieved an accuracy of 0.68. This difference indicates that SVM has better classification capability in handling high-dimensional text data and imbalanced class distributions. This study concludes that SVM is more effective for sentiment analysis of public policy on social media, while Naïve Bayes still offers advantages in computational efficiency and model simplicity. Keywords: Sentiment Analysis, Support Vector Machine, Naïve Bayes, TF–IDF, Social Media X

Item Type:	Thesis (Skripsi)
Uncontrolled Keywords:	Analisis Sentimen, Support Vector Machine, Naïve Bayes, TF–IDF, Media Sosial X==============Sentiment Analysis, Support Vector Machine, Naïve Bayes, TF–IDF, Social Media X
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General) Z Bibliography. Library Science. Information Resources > Z665 Library Science. Information Science Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4050 Electronic information resources Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4450 Databases
Divisions:	Fakultas Sains Dan Teknologi > Sistem Informasi
Depositing User:	Unnamed user with email repository@ulb.ac.id
Date Deposited:	22 Apr 2026 04:24
Last Modified:	22 Apr 2026 04:24
URI:	http://repository.ulb.ac.id/id/eprint/2144

Actions (login required)

View Item