Hidayatuloh, Fahmi Nur (2026) ANALISIS SENTIMEN UNTUK MENGETAHUI PERSPEKTIF MASYARAKAT TERHADAP PROGRAM MAKAN BERGIZI GRATIS DI MEDIA SOSIAL X MENGGUNAKAN METODE NAÏVE BAYES DAN LARGE LANGUAGE MODEL. S1 / D3 thesis, Universitas Kuningan.
ABSTRAK.pdf
Download (1MB)
BAB I.pdf
Download (297kB)
BAB II.pdf
Restricted to Repository staff only
Download (511kB) | Request a copy
BAB III.pdf
Restricted to Repository staff only
Download (1MB) | Request a copy
BAB IV.pdf
Restricted to Repository staff only
Download (463kB) | Request a copy
BAB V.pdf
Download (258kB)
DAFTAR PUSTAKA.pdf
Download (215kB)
LAMPIRAN.pdf
Restricted to Repository staff only
Download (4MB) | Request a copy
Abstract
Program Makan Bergizi Gratis (MBG) merupakan kebijakan pemerintah yang secara implementasinya menimbulkan berbagai respons masyarakat di media sosial X, baik positif, negatif, ataupun netral. Banyaknya opini yang tersebar secara luas menjadi sulit untuk mengetahui kecenderungan masyarakat terhadap Program MBG, sehingga diperlukan pendekatan analisis sentimen secara otomatis. Penelitian ini bertujuan untuk mengetahui kecenderungan sentimen masyarakat terhadap Program MBG serta mengukur kinerja metode Naïve Bayes dengan dukungan pelabelan otomatis menggunakan Large Language Model (LLM). Metode yang dilakukan meliputi pengumpulan data melalui proses crawling tweet berbahasa Indonesia dengan periode waktu Juli 2025- Januari 2026, pre-processing teks, pelabelan otomatis menggunakan LLM, pembobotan TF-IDF, serta klasifikasi dengan metode Naïve Bayes menggunakan sistem penunjang keputusan (SPK). Evaluasi model dilakukan menggunakan metode confusion matrix dengan menghitung nilai akurasi, presisi, recall, dan f1-score. Hasil penelitian menunjukkan bahwa mayoritas memiliki sentimen positif dengan nilai persentase sebesar 55,88%, diikuti sentimen negatif 36,25% dan sentimen netral 7,87%. Temuan ini menunjukkan bahwa persepsi masyarakat terhadap program MBG cenderung positif, sehingga hasil penelitian ini dapat menjadi bahan bagi pemangku kebijakan dalam meningkatkan efektivitas program.
The Free Nutritious Meals Program (MBG) is a government policy that has generated various public responses on the social media platform X, including positive, negative, and neutral sentiments. The large volume of widely distributed opinions makes difficulty to determine overall public perception, thus requiring an automated sentiment analysis approach. This study aims to identify public sentiments toward the MBG program and to evaluate the performance of the Naïve Bayes method supported by automatic labeling using a Large Language Model (LLM). The research methodology includes data collection through crawling Indonesian-language tweets from July 2025 to January 2026, text pre-processing, automatic labeling using an LLM, TF-IDF weighting, and classification using the Naïve Bayes method implemented in a decision support system. Model evaluation is conducted using a confusion matrix by calculating accuracy, precision, recall, and F1-score. The research results show that the majority of sentiments are positive, accounting for 55.88%, followed by negative sentiments at 36.25% and neutral sentiments at 7.87%. These findings indicate that public perception of the MBG program tends to be positive, suggesting that the results can serve as reference for policymakers to improve the program’s effectiveness.
| Item Type: | Thesis (S1 / D3) |
|---|---|
| Uncontrolled Keywords: | Analisis Sentimen, Media Sosial X, Program Makan Bergizi Gratis, Sentiment Analysis, Naïve Bayes, Large Language Model, Social Media X, Free Nutritious Meals Program. |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| Divisions: | Fakultas Ilmu Komputer > S1 Sistem Informasi |
| Depositing User: | S.Kom Fahmi Nur Hidayatuloh |
| Date Deposited: | 15 Sep 2026 02:49 |
| Last Modified: | 15 Sep 2026 02:49 |
| URI: | https://rama.uniku.ac.id/id/eprint/6542 |

