ANALISIS SENTIMEN MENGGUNAKAN MODEL SUPPORT VECTOR MACHINE MENGENAI KERUSAKAN RANGKA ESAF DENGAN PENERAPAN STREAMLIT

Ariyani, Kinanthi Putri (2025) ANALISIS SENTIMEN MENGGUNAKAN MODEL SUPPORT VECTOR MACHINE MENGENAI KERUSAKAN RANGKA ESAF DENGAN PENERAPAN STREAMLIT. Undergraduate thesis, Universitas Pembangunan Nasional Veteran Jawa Timur.

	Text (Cover) 21083010047_Cover.pdf Download (2MB)
	Text (Bab I) 21083010047_Bab I.pdf Download (161kB)
	Text (Bab II) 21083010047_Bab II.pdf Restricted to Registered users only until 17 March 2028. Download (296kB)
	Text (Bab III) 21083010047_Bab III.pdf Restricted to Registered users only until 17 March 2028. Download (248kB)
	Text (Bab IV) 21083010047_Bab IV.pdf Restricted to Registered users only until 17 March 2028. Download (1MB)
	Text (Bab V) 21083010047_Bab V.pdf Download (98kB)
	Text (Daftar Pustaka) 21083010047_Daftar Pustaka.pdf Download (201kB)
	Text (Lampiran) 21083010047_Lampiran.pdf Restricted to Registered users only until 17 March 2028. Download (27kB)

Abstract

The damage to the Enhanced Smart Architecture Frame (eSAF) on Honda motorcycles has raised consumer concerns and become a public spotlight. This study examines public sentiment towards the problem using Support Vector Machine (SVM) and its impact on sales at a dealer in Surabaya. The data used are comments from Twitter social media by dividing the data into two main classification groups, namely positive sentiment and negative sentiment. Based on the analysis results, the majority of 589 public sentiments (59.7%) tend to be negative towards the eSAF frame damage problem, while 397 public sentiments (40.3%) show positive sentiment. The sales results show that negative sentiment on social media does not affect sales figures. The SVM model with Linear and Polynomial kernels gave the best results with 85% accuracy, 85% precision, 85% recall, and 85% f1-score. SVM was chosen because it excels in text classification compared to algorithms such as K-Nearest Neighbors (KNN), C4.5, and Naïve Bayes, and has been applied in various fields such as face detection, bioinformatics, and text processing. This research provides insights for manufacturers to improve product quality, enhance consumer service, and restore public trust. In addition, the use of the SVM algorithm in this sentiment analysis can be a reference for similar research in other fields.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Damaliana, Aviolla Terza	NIDN0002089402	aviolla.terza.sada@upnjatim.ac.id
Thesis advisor	Trimono, Trimono	NIDN0008099501	trimono.stat@upnjatim.ac.id

Subjects:

Q Science > QA Mathematics
Q Science > QA Mathematics > QA76.6 Computer Programming

Divisions:

Faculty of Computer Science > Departemen of Data Science

Depositing User:

Kinanthi Putri Ariyani

Date Deposited:

17 Mar 2025 04:56

Last Modified:

17 Mar 2025 04:56

URI:

https://repository.upnjatim.ac.id/id/eprint/35577