Audyna, Adinda Putri (2024) Sistem Analisis Sentimen Berbasis Aspek pada Media Sosial X terhadap Electric Vehicle (EV) di Indonesia menggunakan INDOBERT dan Machine Learning. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5027201073-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5027201073-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (3MB) | Request a copy |
Abstract
Kementerian Perindustrian menunjukkan bahwa Indonesia sedang mengalami transisi menuju industri ramah lingkungan melalui penggunaan Electric Vehicle (EV) yang didukung oleh Peraturan Presiden Nomor 55 Tahun 2019. EV menjadi salah satu topik yang cukup ramai dibicarakan di Indonesia, khususnya pada media sosial X. Oleh karena itu, penelitian yang menggunakan data dari media sosial X ini bertujuan memanfaatkan teknologi Natural Language Processing (NLP) dengan fokus analisis sentimen berdasarkan beberapa aspek EV (baterai, biaya, infrastruktur, kenyamanan, dan performa) menggunakan metode IndoBERT. Menurut penelitian terdahulu, IndoBERT dinilai lebih efektif dibandingkan dengan metode Machine Learning lainnya. Penggunaan NLP dan IndoBERT diharapkan dapat membuka pandangan masyarakat Indonesia terhadap EV, dengan hasil analisis yang dibandingkan dengan beberapa metode Machine Learning dan Deep Learning. Performa model dievaluasi menggunakan confusion matrix yang mana dari matrix tersebut dapat diketahui nilai accuracy, precision, recall, dan F1 score. Implementasi analisis sentimen berdasarkan beberapa aspek EV akan diaplikasikan dalam bentuk website dashboard sehingga pengguna dapat lebih mudah memperoleh wawasan dan visualisasi data terkait sentimen EV di Indonesia. Dengan demikian, penelitian ini tidak hanya berkontribusi terhadap pemahaman masyarakat dan industri terhadap EV, tetapi juga membuka potensi pengembangan lebih lanjut dalam penerapan teknologi Machine Learning dan Deep Learning untuk analisis sentimen berbasis aspek. Dari hasil penelitian, penggunaan NLP dan IndoBERT menunjukkan performa terbaik dibandingkan metode lainnya, dengan akurasi sentimen mencapai 0,82 dan akurasi aspek mencapai 0,85. Penelitian ini membuat sebuah website dashboard untuk menganalisis sentimen berdasarkan beberapa aspek EV di Indonesia, dan menunjukkan bahwa sentimen negatif dominan dari tahun 2021 hingga 2024. Aspek biaya mendapatkan sentimen positif terbanyak setiap tahunnya, sementara aspek infrastruktur mendapat sentimen negatif terbanyak dari tahun 2021 hingga 2023. Namun aspek baterai mendominasi sentimen negatif pada tahun 2024.
=================================================================================================================================
The Ministry of Industry has indicated that Indonesia is transitioning towards an environmentally friendly industry through the use of Electric Vehicle (EV), supported by Presidential Regulation No. 55 of 2019. EV have become a widely discussed topic in Indonesia, particularly on the social media platform X. Consequently, this research utilizes data from X to leverage Natural Language Processing (NLP) technology, focusing on sentiment analysis based on various aspects of EV (battery, cost, infrastructure, comfort, and performance) using the IndoBERT method. Previous studies have found IndoBERT to be more effective compared to other Machine Learning methods. The use of NLP and IndoBERT aims to broaden the Indonesian public's perspective on EVs, with analysis results compared to several Machine Learning and Deep Learning methods. The model's performance is evaluated using a confusion matrix, from which accuracy, precision, recall, and F1 score values are derived. The implementation of sentiment analysis based on various aspects of EV will be applied in the form of a website dashboard, allowing users to easily gain insights and visualize data related to EV sentiment in Indonesia. This research not only contributes to the public and industry's understanding of EV but also opens up further development potential in the application of Machine Learning and Deep Learning technologies for aspect-based sentiment analysis. The research results indicate that the use of NLP and IndoBERT shows the best performance compared to other methods, with sentiment accuracy reaching 0,82 and aspect accuracy reaching 0,85. This research created a website dashboard to analyze sentiment based on various aspects of EV in Indonesia, revealing that negative sentiment was dominant from 2021 to 2024. The cost aspect received the most positive sentiment each year, while the infrastructure aspect received the most negative sentiment from 2021 to 2023. However, the battery aspect dominated negative sentiment in 2024.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Analisis Sentimen, Analisis Aspek, Electric Vehicle (EV), IndoBERT, Machine Learning, Sentiment Analysis, Aspect Analysis |
| Subjects: | T Technology > T Technology (General) > T11 Technical writing. Scientific Writing T Technology > T Technology (General) > T385 Visualization--Technique T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > T Technology (General) > T58.62 Decision support systems |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information Technology > 59201-(S1) Undergraduate Thesis |
| Depositing User: | Adinda Putri Audyna |
| Date Deposited: | 22 Jul 2024 04:23 |
| Last Modified: | 22 Jul 2024 04:23 |
| URI: | http://repository.its.ac.id/id/eprint/108609 |
Actions (login required)
 |
View Item |