Analisis Sentimen Berbasis Aspek Terhadap Ulasan Pengguna Aplikasi “Gobis Suroboyo”

Amaliana, Dentina Dewi (2026) Analisis Sentimen Berbasis Aspek Terhadap Ulasan Pengguna Aplikasi “Gobis Suroboyo”. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2043221124-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Perkembangan teknologi digital mendorong peningkatan layanan transportasi publik berbasis aplikasi, salah satunya adalah Gobis Suroboyo yang diluncurkan oleh Dinas Perhubungan Kota Surabaya untuk mendukung operasional Suroboyo Bus. Penelitian ini bertujuan mengidentifikasi aspek-aspek utama yang dibahas dalam ulasan pengguna, menganalisis sentimen pada setiap aspek, menentukan prioritas perbaikan, serta mengevaluasi kinerja model dalam memprediksi aspek dan sentimen. Penelitian menerapkan pendekatan Aspect-Based Sentiment Analysis terhadap ulasan yang diperoleh dari Google Play Store, X, dan Instagram. Identifikasi aspek dilakukan menggunakan Latent Dirichlet Allocation, klasifikasi sentimen menggunakan IndoBERT, dan CNN untuk membangun model prediksi aspek dan sentimen. Hasil penelitian menunjukkan bahwa LDA menghasilkan tiga aspek utama, yaitu Manajemen Akun, Transaksi Pembayaran, dan Operasional Transportasi. Analisis sentimen menunjukkan bahwa aspek Manajemen Akun didominasi permasalahan login, pengelolaan saldo, dan top up, aspek Transaksi Pembayaran didominasi kendala pembayaran menggunakan QRIS maupun kartu, sedangkan aspek Operasional Transportasi didominasi keluhan terkait ketepatan waktu, rute, halte, dan layanan bus. Perhitungan NRS menunjukkan seluruh aspek memperoleh nilai negatif. Nilai tersebut menunjukkan bahwa persepsi pengguna terhadap ketiga aspek masih cenderung negatif, dengan aspek Transaksi Pembayaran menjadi prioritas utama perbaikan. Selain itu, model CNN memiliki kemampuan yang baik dalam mengklasifikasikan aspek dan sentimen ulasan pengguna. Hasil penelitian ini diharapkan dapat menjadi dasar bagi pengelola Gobis Suroboyo dalam menetapkan prioritas pengembangan aplikasi guna meningkatkan kualitas layanan transportasi publik berbasis digital.
=====================================================================================================================================
The rapid advancement of digital technology has driven the development of applicationbased public transportation services, including Gobis Suroboyo, an application launched by the Surabaya City Transportation Agency to support Suroboyo Bus operations. This study aims to identify the main aspects discussed in user reviews, analyze sentiment for each aspect, determine improvement priorities, and evaluate the performance of a model for aspect and sentiment prediction. An Aspect-Based Sentiment Analysis approach was applied to user reviews collected from the Google Play Store, X, and Instagram. Aspect identification was performed using Latent Dirichlet Allocation, sentiment classification using IndoBERT, and a Convolutional Neural Network was employed to predict aspects and sentiments. The results identified three main aspects: Account Management, Payment Transactions, and Transportation Operations. The Account Management aspect was mainly associated with login, balance management, and top-up issues, while the Payment Transactions aspect was dominated by QRIS and card payment failures. The Transportation Operations aspect primarily involved complaints regarding punctuality, routes, bus stops, and bus services. Net Reputation Score analysis indicated negative scores across all aspects, with Payment Transactions identified as the highest priority for improvement. In addition, the CNN model demonstrated good performance in classifying review aspects and sentiments. These findings provide insights for Gobis Suroboyo management in prioritizing application improvements to enhance the quality of digital public transportation services.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	CNN, Gobis Suroboyo, IndoBERT, LDA, Net Reputation Score, CNN, Gobis Suroboyo, IndoBERT, LDA, Net Reputation Score
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science)
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Dentina Dewi Amaliana
Date Deposited:	31 Jul 2026 06:57
Last Modified:	31 Jul 2026 06:57
URI:	http://repository.its.ac.id/id/eprint/140781

Actions (login required)

View Item