Hamdani, Muhammad Aqsa (2024) Analisis sentimen, pemodelan topik, dan visualisasi komentar konter kuliner YouTube terhadap "WNK" Restoran dengan menggunakan Vader untuk mendukung bisnis kuliner. Masters thesis, Universitas Pelita Harapan.
![Title [thumbnail of Title]](http://repository.uph.edu/style/images/fileicons/text.png)
Cover_MUHAMMAD AQSA HAMDAN 1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (167kB)
![Abstract [thumbnail of Abstract]](http://repository.uph.edu/style/images/fileicons/text.png)
Abstrak_watermark_MUHAMMAD AQSA HAMDAN.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (399kB)
![ToC [thumbnail of ToC]](http://repository.uph.edu/style/images/fileicons/text.png)
Daftar isi_watermark_MUHAMMAD AQSA HAMDAN.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (558kB)
![Chapter [thumbnail of Chapter]](http://repository.uph.edu/style/images/fileicons/text.png)
Bab 1_watermark_MUHAMMAD AQSA HAMDAN.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (919kB)
![Chapter 2 [thumbnail of Chapter 2]](http://repository.uph.edu/style/images/fileicons/text.png)
Bab 2_watermark_MUHAMMAD AQSA HAMDAN.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (823kB)
![Chapter 3 [thumbnail of Chapter 3]](http://repository.uph.edu/style/images/fileicons/text.png)
Bab 3_watermark_MUHAMMAD AQSA HAMDAN.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (1MB)
![Chapter 4 [thumbnail of Chapter 4]](http://repository.uph.edu/style/images/fileicons/text.png)
Bab 4_watermark_MUHAMMAD AQSA HAMDAN.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (2MB)
![Chapter 5 [thumbnail of Chapter 5]](http://repository.uph.edu/style/images/fileicons/text.png)
Bab 5_watermark_MUHAMMAD AQSA HAMDAN.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (339kB)
![Bibliography [thumbnail of Bibliography]](http://repository.uph.edu/style/images/fileicons/text.png)
Daftar Pustaka_watermark_MUHAMMAD AQSA HAMDAN.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (539kB)
![Appendices [thumbnail of Appendices]](http://repository.uph.edu/style/images/fileicons/text.png)
Lampiran_watermark_MUHAMMAD AQSA HAMDAN.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (4MB)
Abstract
Comments on YouTube culinary content in food vlogger videos are a
valuable source of information to support the development of culinary businesses.
Sentiment analysis offers an automated solution to categorize reviews into positive,
neutral, or negative sentiments. This study applies the Valence Aware Dictionary
for Sentiment Reasoning (VADER) model to Indonesian comments discussing
culinary content related to the "WNK" restaurant, by modifying VADER to be able
to identify sentiment polarity in Indonesian more accurately.
This study uses two sentiment analysis models, namely VADER Default
and VADER Custom, which modify the English-based VADER polarity lexicon to
classify Indonesian text sentiment. Comment data from three YouTube videos are
processed through pre-processing stages, such as text cleaning, tokenization,
normalization, and stopword removal. Model performance is evaluated using
accuracy, precision, recall, and F1-score, while word cloud and N-gram analysis
are used to identify main themes and dominant phrases.
The results show that the VADER Custom Indonesian model significantly
improves the performance of Indonesian text sentiment analysis. This improvement
is proven through the polarity distribution value of 5,20% compared to the VADER
Default model of 2,24%, as well as the evaluation of the Precision value of 76,94%,
Recall 76,48%, and F1 Score 76,68%. The use of topic analysis and visualization
provides useful insights for business actors to improve service quality, with a focus
on price, food quality, and cleanliness. / Komentar pada konten kuliner YouTube di video food vlogger merupakan sumber informasi yang berharga dalam mendukung pengembangan bisnis kuliner. Analisis sentimen menawarkan solusi otomatis untuk mengkategorikan ulasan ke dalam sentimen positif, netral, atau negatif. Penelitian ini menerapkan model Valence Aware Dictionary for Sentiment Reasoning (VADER) pada komentar berbahasa Indonesia yang membahas konten kuliner terkait restoran "WNK", dengan melakukan modifikasi pada VADER agar mampu mengidentifikasi polaritas sentimen dalam Bahasa Indonesia secara lebih akurat.
Penelitian ini menggunakan dua model analisis sentimen, yaitu VADER Default dan VADER Custom, yang memodifikasi leksikon polaritas VADER berbasis bahasa Inggris untuk mengklasifikasikan sentimen teks Bahasa Indonesia. Data komentar dari tiga video YouTube diproses melalui tahapan pra-pemrosesan, seperti pembersihan teks, tokenisasi, normalisasi, dan penghapusan stopword. Kinerja model dievaluasi menggunakan akurasi, precision, recall, dan F1-score, sementara analisis word cloud dan N-gram digunakan untuk mengidentifikasi tema utama dan frasa dominan.
Hasil penelitian menunjukkan bahwa model VADER Custom Bahasa Indonesia secara signifikan meningkatkan kinerja analisis sentimen teks Bahasa Indonesia. Peningkatan ini dibuktikan melalui nilai distribusi polaritas sebanyak 5,20% dibandingkan model VADER Default 2,24%, serta evaluasi nilai Precision 76,94%, Recall 76,48%, dan F1 Score 76,68%. Pemanfaatan analisis topik dan visualisasi memberikan wawasan yang berguna bagi pelaku bisnis untuk meningkatkan kualitas layanan, dengan fokus pada harga, kualitas makanan, dan kebersihan.
Item Type: | Thesis (Masters) |
---|---|
Creators: | Creators NIM Email ORCID Hamdani, Muhammad Aqsa NIM01679230001 UNSPECIFIED UNSPECIFIED |
Contributors: | Contribution Contributors NIDN/NIDK Email Thesis advisor Tjahyadi, Hendra NIDN0410076901 UNSPECIFIED |
Uncontrolled Keywords: | Analisis Sentiment ; YouTube ; VADER Bahasa Indonesia ; Bisnis Kuliner |
Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7885-7895 Computer engineering. Computer hardware |
Divisions: | University Subject > Current > Faculty/School - UPH Karawaci > School of Information Science and Technology > Master of Informatics Current > Faculty/School - UPH Karawaci > School of Information Science and Technology > Master of Informatics |
Depositing User: | Phillips Iman Heri Wahyudi |
Date Deposited: | 26 Feb 2025 07:19 |
Last Modified: | 26 Feb 2025 07:19 |
URI: | http://repository.uph.edu/id/eprint/67328 |