Analisis Sentimen Ulasan Aplikasi Mobile Banking (Brimo) Menggunakan Algoritma Support Vector Machine (Svm) Dan Naives Bayes

Abstrak

Abstrak
Perkembangan layanan mobile banking mendorong meningkatnya jumlah ulasan pengguna pada platform digital seperti Google Play Store. Ulasan tersebut mengandung informasi penting mengenai tingkat kepuasan pengguna terhadap kualitas layanan aplikasi BRImo. Namun, jumlah data ulasan yang besar dan berbentuk teks tidak terstruktur menyebabkan proses analisis secara manual menjadi kurang efektif. Penelitian ini bertujuan untuk melakukan analisis sentimen terhadap ulasan aplikasi BRImo menggunakan algoritma SVM dan Naive Bayes guna membandingkan performa kedua metode dalam klasifikasi sentimen pengguna. Metode penelitian dilakukan melalui beberapa tahapan, yaitu pengumpulan data, preprocessing data, feature extraction menggunakan TF-IDF, pembagian data latih dan data uji, proses pelatihan model, serta evaluasi performa model. Tahap preprocessing meliputi case folding, tokenizing, filtering, normalisasi, dan stemming untuk meningkatkan kualitas data teks sebelum proses klasifikasi dilakukan. Algoritma SVM digunakan karena kemampuannya dalam menangani data teks berdimensi tinggi dengan tingkat akurasi yang baik, sedangkan Naive Bayes digunakan karena memiliki proses komputasi yang sederhana dan efisien dalam klasifikasi teks. Evaluasi model dilakukan menggunakan confusion matrix dengan metrik accuracy, precision, recall, dan F1-score. Hasil penelitian diharapkan mampu menunjukkan algoritma yang memiliki performa terbaik dalam mengklasifikasikan sentimen ulasan pengguna aplikasi BRImo. Selain itu, penelitian ini diharapkan dapat memberikan kontribusi dalam pengembangan analisis sentimen berbasis machine learning serta menjadi bahan evaluasi bagi pengembang aplikasi dalam meningkatkan kualitas layanan mobile banking.
Kata kunci: Analisis Sentimen, BRImo, Support Vector Machine, Naive Bayes, Machine Learning
Abstract
The development of mobile banking services has increased the number of user reviews on digital platforms such as Google Play Store. These reviews contain important information regarding user satisfaction with the quality of the BRImo application service. However, the large amount of review data in unstructured text form makes manual analysis less effective. This study aims to conduct sentiment analysis on BRImo application reviews using SVM and Naive Bayes algorithms to compare the performance of both methods in classifying user sentiment. The research method consists of several stages, including data collection, data preprocessing, feature extraction using TF-IDF, training and testing data splitting, model training, and model performance evaluation. The preprocessing stage includes case folding, tokenizing, filtering, normalization, and stemming to improve the quality of text data before classification. The SVM algorithm is used due to its capability in handling high-dimensional text data with good accuracy, while Naive Bayes is applied because of its simple and efficient computational process in text classification. Model evaluation is carried out using a confusion matrix with accuracy, precision, recall, and F1-score metrics. The results of this study are expected to identify the algorithm with the best performance in classifying sentiment from BRImo user reviews. In addition, this research is expected to contribute to the development of machine learning-based sentiment analysis and serve as evaluation material for application developers in improving mobile banking service quality.
Keyword: Sentiment Analysis, BRImo, Support Vector Machine, Naive Bayes, Machine Learning