PERBANDINGAN DUA EKSTRASI FITUR UNTUK ANALISIS SENTIMEN MENGGUNAKAN NAIVE BAYES PADA MEDIA SOSIAL STUDI KASUS PEJABAT PUBLIK LULUS S3 KURANG DARI DUA TAHUN

SAPUTRA, FERDIANSYAH (2025) PERBANDINGAN DUA EKSTRASI FITUR UNTUK ANALISIS SENTIMEN MENGGUNAKAN NAIVE BAYES PADA MEDIA SOSIAL STUDI KASUS PEJABAT PUBLIK LULUS S3 KURANG DARI DUA TAHUN. S1 thesis, Universitas Mercu Buana Jakarta - Menteng.

[img] Text (COVER)
41521010207-FERDIANSYAH SAPUTRA- 01 COVER - FERDIANSYAH SAPUTRA.pdf
Download (205kB)
[img] Text (BAB I)
41521010207-FERDIANSYAH SAPUTRA-02 BAB 1 - FERDIANSYAH SAPUTRA.pdf
Restricted to Registered users only
Download (45kB)
[img] Text (BAB II)
41521010207-FERDIANSYAH SAPUTRA-03 BAB 2 - FERDIANSYAH SAPUTRA.pdf
Restricted to Registered users only
Download (117kB)
[img] Text (BAB III)
41521010207-FERDIANSYAH SAPUTRA-04 BAB 3 - FERDIANSYAH SAPUTRA.pdf
Restricted to Registered users only
Download (107kB)
[img] Text (BAB IV)
41521010207-FERDIANSYAH SAPUTRA-05 BAB 4 - FERDIANSYAH SAPUTRA.pdf
Restricted to Registered users only
Download (511kB)
[img] Text (BAB V)
41521010207-FERDIANSYAH SAPUTRA-06 BAB 5 - FERDIANSYAH SAPUTRA.pdf
Restricted to Registered users only
Download (35kB)
[img] Text (DAFTAR PUSTAKA)
41521010207-FERDIANSYAH SAPUTRA-08 DAFTAR PUSTAKA - FERDIANSYAH SAPUTRA.pdf
Restricted to Registered users only
Download (57kB)
[img] Text (LAMPIRAN)
41521010207-FERDIANSYAH SAPUTRA-09 LAMPIRAN - FERDIANSYAH SAPUTRA.pdf
Restricted to Registered users only
Download (714kB)

Abstract

Media sosial telah menjadi saluran utama masyarakat dalam menyampaikan opini terhadap isu-isu publik, termasuk pencapaian akademik pejabat negara. Salah satu isu yang ramai diperbincangkan adalah disertasi Menteri Energi dan Sumber Daya Mineral yang diselesaikan dalam waktu kurang dari dua tahun. Penelitian ini bertujuan untuk menganalisis sentimen publik terhadap disertasi Menteri Energi dan Sumber Daya Mineral, yang diselesaikan dalam waktu 1 tahun 8 bulan. Data yang digunakan dalam penelitian ini diambil dari platform media sosial Twitter, dengan total 2100 komentar yang dikumpulkan menggunakan teknik crawling berdasarkan kata kunci "Disertasi Menteri Energi Dan Sumber Daya Mineral". Metode yang diterapkan dalam penelitian ini adalah analisis sentimen menggunakan algoritma Naive Bayes, dengan perbandingan dua metode ekstraksi fitur, yaitu Term Frequency-Inverse Document Frequency (TF-IDF) dan Bag of Words (BoW). Proses penelitian meliputi tahapan pengumpulan data, preprocessing, pelabelan sentimen, dan evaluasi model. Hasil evaluasi menunjukkan bahwa model Naive Bayes dengan metode BoW memberikan akurasi yang lebih tinggi (83.01%) dibandingkan dengan TF-IDF (81.76%), serta menunjukkan performa yang lebih stabil dalam klasifikasi sentimen positif dan negatif. Penelitian ini diharapkan dapat memberikan wawasan yang lebih dalam mengenai persepsi masyarakat terhadap prestasi akademis pejabat publik dan implikasinya terhadap kepercayaan masyarakat. Selain itu, hasil penelitian ini dapat menjadi referensi bagi penelitian selanjutnya dalam bidang analisis sentimen di media sosial. Social media has become the main channel for people to express their opinions on public issues, including the academic achievements of state officials. One of the issues that is widely discussed is the dissertation of the Minister of Energy and Mineral Resources which was completed in less than two years. This study aims to analyze public sentiment towards the dissertation of the Minister of Energy and Mineral Resources, which was completed in 1 year and 8 months. The data used in this study were taken from the social media platform Twitter, with a total of 2100 comments collected using the crawling technique based on the keyword "Disertasi Menteri Energi Dan Sumber Daya Mineral". The method applied in this study is sentiment analysis using the Naive Bayes algorithm, with a comparison of two feature extraction methods, namely Term Frequency-Inverse Document Frequency (TF-IDF) and Bag of Words (BoW). The research process includes the stages of data collection, preprocessing, sentiment labeling, and model evaluation. The evaluation results show that the Naive Bayes model with the BoW method provides higher accuracy (83.01%) compared to TF-IDF (81.76%), and shows more stable performance in the classification of positive and negative sentiments. This study is expected to provide deeper insight into public perceptions of public officials' academic achievements and their implications for public trust. In addition, the results of this study can be a reference for further research in the field of sentiment analysis on social media.

Item Type: Thesis (S1)
NIM/NIDN Creators: 41521010207
Uncontrolled Keywords: Analisis Sentimen, Naïve Bayes, TF-IDF,BoW,BERT. Sentiment Analysis, Naïve Bayes, TF-IDF,BoW,BERT.
Subjects: 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 004 Data Processing, Computer Science/Pemrosesan Data, Ilmu Komputer, Teknik Informatika
Divisions: Fakultas Ilmu Komputer > Informatika
Depositing User: NAIMAH NUR ISLAMIDIYANAH
Date Deposited: 12 Aug 2025 06:58
Last Modified: 12 Aug 2025 06:58
URI: http://repository.mercubuana.ac.id/id/eprint/96784

Actions (login required)

View Item View Item
UMB Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.