Analisis Sentimen Masyarakat Terhadap Layanan Pinjaman Online Menggunakan Metode Support Vector Machine (SVM)

Firmansyah, Tegar Maulana (2026) Analisis Sentimen Masyarakat Terhadap Layanan Pinjaman Online Menggunakan Metode Support Vector Machine (SVM). Undergraduate thesis, Politeknik Negeri Jember.

	Text (Abstract) ABSTRACT.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (143kB)
	Text (Bab 1 Pendahuluan) BAB 1 PENDAHULUAN.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (170kB)
	Text (Daftar Pustaka) DAFTAR PUSTAKA.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (155kB)
	Text (Laporan Lengkap) 1REVISI LAPORAN LENGKAP.pdf Restricted to Registered users only Download (3MB) | Request a copy

Abstract

Online lending services offer financial convenience but also raise various public complaints such as high interest rates and unethical collection practices, therefore, this study aims to analyze user sentiment toward the Easycash application on the Google Play Store using the Support Vector Machine (SVM) method and to evaluate the performance of the resulting model. Data were collected through a two-stage crawling process, resulting in a total of 2,550 reviews, followed by preprocessing steps including case folding, cleaning (removal of punctuation, numbers, and symbols), tokenizing, slang word normalization, stopword removal, and stemming to produce clean and structured text data. Furthermore, term weighting was performed using the TF-IDF method to transform text into numerical representations before classification using the SVM algorithm into positive, negative, and neutral sentiment categories. SVM combined with TF-IDF is commonly used and effective for sentiment analysis tasks. The results show that the SVM model performs well in sentiment classification, particularly in identifying positive sentiment, with the best testing scenario (second scenario) achieving an accuracy of 87.33% with 655 correct predictions out of 750 testing data. however, the model still faces challenges in distinguishing between negative and neutral sentiments due to similarities in linguistic characteristics. This ix study is expected to serve as an evaluation tool for online lending service providers and as a reference for regulators such as OJK in understanding public perception.

Actions (login required)

View Item