ANALISIS SENTIMEN TIME-SERIES REPUTASI GREEN SM PASCA KECELAKAAN STASIUN BEKASI TIMUR: NAIVE BAYES VS BI-GRU

KHOLIFAH, UMI NUR (2026) ANALISIS SENTIMEN TIME-SERIES REPUTASI GREEN SM PASCA KECELAKAAN STASIUN BEKASI TIMUR: NAIVE BAYES VS BI-GRU. S1 thesis, Universitas Mercu Buana Jakarta.

[img]
Preview
Text (HAL COVER)
COVER.pdf

Download (431kB) | Preview
[img] Text (BAB I)
BAB 1.pdf
Restricted to Registered users only

Download (47kB)
[img] Text (BAB II)
BAB 2.pdf
Restricted to Registered users only

Download (319kB)
[img] Text (BAB III)
BAB 3.pdf
Restricted to Registered users only

Download (469kB)
[img] Text (BAB IV)
BAB 4.pdf
Restricted to Registered users only

Download (1MB)
[img] Text (BAB V)
BAB 5.pdf
Restricted to Registered users only

Download (29kB)
[img] Text (DAFTAR PUSTAKA)
DAFTAR PUSTAKA.pdf
Restricted to Registered users only

Download (168kB)
[img] Text (LAMPIRAN)
LAMPIRAN.pdf
Restricted to Registered users only

Download (521kB)

Abstract

Social media has become an important space for the public to express opinions about various events, including events that may affect a company's reputation. One such event was the accident involving a Green SM vehicle at Bekasi Timur Station on April 27, 2026. This study aims to analyze changes in public sentiment toward Green SM before and after the accident using a time-series approach and to compare the performance of Complement Naïve Bayes (CNB), Multinomial Naïve Bayes (MNB), and Bidirectional Gated Recurrent Unit (Bi-GRU). The dataset consists of 3,339 Indonesian-language tweets collected from Twitter/X during the period from December 12, 2024, to July 27, 2026. The data underwent preprocessing and were then classified using MNB, CNB, and Bi-GRU. The results show that Bi-GRU Optimized (Voting) achieved the best performance, with an accuracy of 82.75% and an F1-Macro score of 0.7762 on the 80:20 split, and was therefore used to label the data for the time-series analysis. The analysis revealed a significant change in sentiment following the accident, with negative sentiment increasing from 35.2% to 72.8%, while positive sentiment decreased from 33.4% to 0.8%. The Mann-Whitney test produced a p-value of 0.0000, indicating a significant difference in sentiment before and after the accident. The time-series analysis also showed a decline in the Net Sentiment Score (NSS) toward negative values following the incident. Kata kunci: Sentiment Analysis, Twitter/X, Green SM, Naïve Bayes, Bi-GRU, Time Series. Perkembangan media sosial menjadikan Twitter/X sebagai ruang bagi masyarakat untuk menyampaikan opini terhadap berbagai peristiwa, termasuk peristiwa yang dapat memengaruhi reputasi perusahaan. Salah satunya adalah kecelakaan yang melibatkan armada Green SM di Stasiun Bekasi Timur pada 27 April 2026. Penelitian ini bertujuan menganalisis perubahan sentimen publik terhadap Green SM sebelum dan sesudah kecelakaan menggunakan pendekatan time-series serta membandingkan kinerja Complement Naïve Bayes (CNB), Multinomial Naïve Bayes (MNB) dan Bidirectional Gated Recurrent Unit (Bi-GRU). Data penelitian terdiri dari 3.339 tweet berbahasa Indonesia yang dikumpulkan dari Twitter/X pada periode 12 Desember 2024 hingga 27 Juli 2026. Data melalui tahap preprocessing, kemudian diklasifikasikan menggunakan MNB, CNB dan Bi-GRU. Hasil pengujian menunjukkan bahwa Bi-GRU Optimized (Voting) memberikan performa terbaik dengan akurasi 82,75% dan F1-Macro 0,7762 pada split 80:20, sehingga digunakan untuk pelabelan data pada analisis time-series. Hasil analisis menunjukkan perubahan sentimen yang signifikan setelah kecelakaan, dengan sentimen negatif meningkat dari 35,2% menjadi 72,8%, sedangkan sentimen positif menurun dari 33,4% menjadi 0,8%. Uji Mann-Whitney menghasilkan nilai p=0,0000 yang menunjukkan adanya perbedaan sentimen yang signifikan sebelum dan sesudah kecelakaan. Analisis time-series juga menunjukkan penurunan Net Sentiment Score (NSS) ke arah negatif setelah kejadian. Kata kunci: Analisis Sentimen, Twitter/X, Green SM, Naïve Bayes, Bi-GRU, Deret Waktu.

Item Type: Thesis (S1)
NIM/NIDN Creators: 41521120007
Uncontrolled Keywords: Analisis Sentimen, Twitter/X, Green SM, Naïve Bayes, Bi-GRU, Deret Waktu.
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
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 > 006 Special Computer Methods/Metode Komputer Tertentu > 006.7 Multimedia Systems/Sistem-sistem Multimedia > 006.75 Social Multimedia/Multimedia Social
100 Philosophy and Psychology/Filsafat dan Psikologi > 150 Psychology/Psikologi > 155 Differential and Developmental Psychology/Psikologi Diferensial dan Psikologi Perkembangan > 155.9 Environmental Psychology/Psikologi Lingkungan > 155.93 Influence of Specific Situations/Pengaruh Situasi Tertentu > 155.936 Accidents/Kecelakaan
300 Social Science/Ilmu-ilmu Sosial > 300. Social Science/Ilmu-ilmu Sosial > 303 Social Process/Proses Sosial > 303.3 Coordination and Control/Koordinasi dan Kontrol > 303.38 Public Opinion/Opini Publik
500 Natural Science and Mathematics/Ilmu-ilmu Alam dan Matematika > 510 Mathematics/Matematika > 518 Numerical Analysis/Analisis Numerik, Analisa Numerik > 518.1 Algorithms/Algoritma
Divisions: Fakultas Ilmu Komputer > Informatika
Depositing User: khalimah
Date Deposited: 07 Sep 2026 14:07
Last Modified: 07 Sep 2026 14:07
URI: http://repository.mercubuana.ac.id/id/eprint/103680

Actions (login required)

View Item View Item