Laksana, Dimaz Putra (2024) ANALISIS SENTIMEN TERHADAP INSTANSI PENGELOLA PANGAN INDONESIA PADA TWITTER MENGGUNAKAN KOMBINASI MODEL BERT DAN NEURAL NETWORKS. S1 thesis, Universitas Mercu Buana - Menteng.
![]() |
Text (COVER)
41520110091-Dimaz Putra Laksana-01 Cover - DIMAZ PUTRA LAKSANA.pdf Download (252kB) |
![]() |
Text (BAB I)
41520110091-Dimaz Putra Laksana-03 Bab 1 - DIMAZ PUTRA LAKSANA.pdf Restricted to Registered users only Download (43kB) |
![]() |
Text (BAB II)
41520110091-Dimaz Putra Laksana-04 Bab 2 - DIMAZ PUTRA LAKSANA.pdf Restricted to Registered users only Download (292kB) |
![]() |
Text (BAB III)
41520110091-Dimaz Putra Laksana-05 Bab 3 - DIMAZ PUTRA LAKSANA.pdf Restricted to Registered users only Download (189kB) |
![]() |
Text (BAB IV)
41520110091-Dimaz Putra Laksana-06 Bab 4 - DIMAZ PUTRA LAKSANA.pdf Restricted to Registered users only Download (4MB) |
![]() |
Text (BAB V)
41520110091-Dimaz Putra Laksana-07 Bab 5 - DIMAZ PUTRA LAKSANA.pdf Restricted to Registered users only Download (35kB) |
![]() |
Text (DAFATR PUSTAKA)
41520110091-Dimaz Putra Laksana-09 Daftar Pustaka - DIMAZ PUTRA LAKSANA.pdf Restricted to Registered users only Download (146kB) |
![]() |
Text (LAMPIRAN)
41520110091-Dimaz Putra Laksana-10 Lampiran - DIMAZ PUTRA LAKSANA.pdf Restricted to Registered users only Download (1MB) |
Abstract
Dalam era digital, Twitter menjadi media utama bagi Instansi Pengelola Pangan Indonesia untuk berkomunikasi dengan masyarakat. Penelitian ini bertujuan menganalisis sentimen cuitan mengenai instansi tersebut menggunakan model BERT yang dikombinasikan dengan model neural networks seperti CNN, LSTM, BiLSTM, GRU, dan BiGRU. Data cuitan dikumpulkan menggunakan twscrape dan dianalisis melalui tahapan pra-pemrosesan, pelatihan, dan evaluasi model berdasarkan akurasi, presisi, recall, dan F1-score. Studi ini membandingkan efektivitas berbagai kombinasi model dalam analisis sentimen untuk mengidentifikasi yang paling optimal. Hasil penelitian diharapkan memberikan wawasan tentang persepsi publik dan strategi komunikasi yang efektif bagi instansi melalui media sosial. Penelitian ini juga menyumbang pada pengembangan metode analisis sentimen dalam bahasa Indonesia. In the digital era, Twitter has become a primary platform for the Indonesian Food Agency to communicate with the public. This study aims to analyze sentiment in tweets related to the agency using the BERT model combined with neural network models such as CNN, LSTM, BiLSTM, GRU, and BiGRU. Tweet data is collected utilizing twscrape, a Python library that emulates human behavior in accessing Twitter. The research process involves data preprocessing, model training, and model evaluation based on accuracy, precision, recall, and F1-score. This study compares the effectiveness of various model combinations in sentiment analysis to identify the optimal one. The results are expected to provide insights into public perception and effective communication strategies for the agency on social media. This study also contributes to the development of sentiment analysis methods for the Indonesian language.
Item Type: | Thesis (S1) |
---|---|
NIM/NIDN Creators: | 41520110091 |
Uncontrolled Keywords: | Analisis Sentimen, Instansi Pengelola Pangan Indonesia, twscrape, BERT, Neural Networks Sentimen Analysis, Indonesian Food Management Agency, twscrape, BERT, Neural Networks |
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: | SILMI KAFFA MARISKA |
Date Deposited: | 24 Aug 2024 03:55 |
Last Modified: | 24 Aug 2024 03:55 |
URI: | http://repository.mercubuana.ac.id/id/eprint/90686 |
Actions (login required)
![]() |
View Item |