Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/155002
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorSartono, Bagus
dc.contributor.advisorSilvianti, Pika
dc.contributor.authorPartawijaya, Herdian
dc.date.accessioned2024-07-29T07:01:20Z
dc.date.available2024-07-29T07:01:20Z
dc.date.issued2024
dc.identifier.urihttp://repository.ipb.ac.id/handle/123456789/155002
dc.description.abstractPenambahan volume kendaraan bermotor terutama kendaraan pribadi dapat menimbulkan berbagai dampak negatif. Saat ini pemerintah menghadirkan fasilitas transportasi publik terbaru yaitu kereta cepat Jakarta – Bandung sebagai salah satu solusi untuk mengatasi hal tersebut. Penelitian ini bertujuan untuk mengklasifikasikan sentimen masyarakat terhadap layanan kereta cepat Jakarta – Bandung dan mengidentifikasi performa penggunaan metode Multiclass Support Vector Machine, serta mengidentifikasi topik dalam sentimen masyarakat terhadap layanan kereta cepat Jakarta – Bandung. Hanya sebanyak 40% data yang diberikan label sentimen secara manual yang digunakan dalam proses pelatihan model dan sisanya akan diprediksi menggunakan model terbaik. Hasil dari penelitian ini diperoleh model terbaik adalah algoritma One-Against-All kernel sigmoid dengan hyperparameter C = 1000, dan gamma = 0,0001. Performa yang dihasilkan yaitu akurasi sebesar 74,8% dan macro average f1-score sebesar 74,6%. Pelabelan menggunakan model terbaik terhadap 14.255 data diperoleh sentimen positif sebanyak 3.715, netral sebanyak 5.288, dan negatif sebanyak 5.252. Hasil clustering sentimen positif menunjukkan berbagai pujian dari masyarakat, serta rasa bangga karena merupakan kereta cepat pertama di Asia Tenggara. Masyarakat juga merasa kagum dengan kecepatan dan efisiensi waktu perjalanan kereta cepat. Sedangkan sentimen negatif menunjukkan kekecewaan biaya kereta cepat yang menggunakan APBN dan kritik bahwa kereta cepat tidak terlalu penting karena tidak ada urgensinya apalagi ibu kota Indonesia akan dipindahkan. Masyarakat juga mengkritik biaya kereta cepat yang terus membengkak hingga ratusan triliun dan stasiunnya yang hanya sampai di Padalarang.
dc.description.abstractThe increase in the volume of motorized vehicles, especially private vehicles, can lead to various negative impacts. Currently, the government has introduced a new public transportation facility, the Jakarta-Bandung high-speed train, as a solution to address these issues. This research aims to classify public sentiment towards the Jakarta - Bandung high-speed train service and identify the performance of the Multiclass Support Vector Machine method, as well as identifying topics in public sentiment towards the Jakarta - Bandung high-speed train service. Only 40% of the data manually labeled with sentiment was used in the model training process, with the remainder being predicted using the best model. The results of this research show that the best model is the One-Against-All algorithm with a sigmoid kernel, using hyperparameters C = 1000 and gamma = 0.0001. The performance achieved was an accuracy of 74.8% and a macro average F1-score of 74.6%. Labeling the remaining 14,255 data points using the best model resulted in 3,715 positive sentiments, 5,288 neutral sentiments, and 5,252 negative sentiments. Clustering of the positive sentiments revealed various praises from the public, including pride in having the first high-speed train in Southeast Asia and admiration for its speed and travel time efficiency. On the other hand, negative sentiments highlighted disappointment over the high-speed train's use of state budget funds, criticisms about its perceived lack of necessity given the upcoming relocation of Indonesia's capital, and concerns over the escalating costs exceeding hundreds of trillions, with the train only reaching Padalarang station.
dc.description.sponsorship
dc.language.isoid
dc.publisherIPB Universityid
dc.titleAnalisis Sentimen Masyarakat Terhadap Layanan Kereta Cepat Jakarta – Bandung Menggunakan Metode Multiclass Support Vector Machine.id
dc.title.alternativeSentiment Analysis of the Public Towards the Jakarta-Bandung High-Speed Train Service Using the Multiclass Support Vector Machine Method
dc.typeSkripsi
dc.subject.keywordhigh speed railid
dc.subject.keywordk-means clusteringid
dc.subject.keywordmulticlass support vector machineid
dc.subject.keywordsentiment analysisid
dc.subject.keywordtwitterid
Appears in Collections:UT - Statistics and Data Sciences

Files in This Item:
File Description SizeFormat 
cover_G1401201022_e178e3a0b4b044b387e9ca0b9c921565.pdfCover1.1 MBAdobe PDFView/Open
fulltext_G1401201022_5ad1e965cacd481c82800f20d2cc139f.pdf
  Restricted Access
Fulltext8.09 MBAdobe PDFView/Open
lampiran_G1401201022_ed5ccf1d23404daf9baf2c2db7d26a10.pdf
  Restricted Access
Lampiran735.18 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.