Tambunan, Rafael Asi Kristanto (2025) Analisis Sentimen Pada Ulasan Pengguna Video Resep Makanan Pada Saluran Youtube Berbahasa Inggris. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5025201168-Non_Degree.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5025201168-Non_Degree.pdf Restricted to Repository staff only Download (4MB) | Request a copy |
|
Text
5025201168-Undergraduated_thesis.pdf Restricted to Repository staff only until 1 April 2027. Download (4MB) | Request a copy |
Abstract
Dalam era digital yang didominasi oleh platform media sosial, analisis sentimen terhadap konten video, khususnya resep masakan di YouTube, menjadi penting untuk memahami pandangan pengguna dan meningkatkan pengalaman pencarian resep. Penelitian ini menggunakan dataset berjumlah 7.221 ulasan dari video resep masakan berbahasa Inggris di saluran YouTube "Kay's Cooking." Setelah melalui proses pembersihan, data yang digunakan berjumlah 6.560 ulasan, yang dilabeli secara manual ke dalam tiga kategori sentimen: positif, negatif, dan netral. Penelitian ini mengusulkan tiga metode klasifikasi, yaitu Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM), dan Bidirectional LSTM (BiLSTM), dengan perhatian khusus pada BiLSTM yang dilengkapi dengan attention layer. Evaluasi model dilakukan menggunakan metrik precision, recall, F1-score, dan akurasi. Hasil menunjukkan bahwa BiLSTM dengan attention layer mencapai performa terbaik dengan akurasi rata-rata sebesar 74,13%, diikuti oleh LSTM dengan akurasi 73,77%, dan GRU dengan akurasi 73,38%. BiLSTM juga unggul dalam mengenali sentimen negatif dan positif dibandingkan model lainnya.
==================================================================================================================================
In the digital era dominated by social media platforms, sentiment analysis of video content, particularly cooking recipes on YouTube, plays a crucial role in understanding user perspectives and enhancing recipe search experiences. This study utilized a dataset of 7,221 reviews from English-language cooking recipe videos on the YouTube channel "Kay's Cooking." After a data-cleaning process, the dataset was refined to 6,560 reviews, which were manually labeled into three sentiment categories: positive, negative, and neutral. This research proposed three classification methods: Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM), and Bidirectional LSTM (BiLSTM), with a specific focus on BiLSTM enhanced by an attention layer. Model evaluation was conducted using precision, recall, F1-score, and accuracy metrics. The results revealed that BiLSTM with an attention layer achieved the best performance, with an average accuracy of 74.13%, followed by LSTM at 73.77% and GRU at 73.38%. BiLSTM also excelled in identifying negative and positive sentiments compared to the other models.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Deep Learning, Food Recipe, Sentiment Analysis, Youtube, Analisis Sentimen, Deep Learning, Resep Masakan, Youtube. |
| Subjects: | T Technology > T Technology (General) T Technology > T Technology (General) > T57.5 Data Processing |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Rafael Asi Kristanto Tambunan |
| Date Deposited: | 03 Feb 2025 02:44 |
| Last Modified: | 03 Feb 2025 02:44 |
| URI: | http://repository.its.ac.id/id/eprint/117567 |
Actions (login required)
 |
View Item |