Pramadani, Indri (2025) Analisis Sentimen Publik Terhadap Kinerja Pelayanan Pt. Bank Rakyat Indonesia (persero) Tbk. Menggunakan Algoritma Support Vector Machine (svm). Other thesis, Universitas Islam Riau.

Text 203510618.pdf - Submitted Version Restricted to Registered users only Download (3MB) | Request a copy

Abstract

The service performance of PT Bank Rakyat Indonesia (Persero) Tbk is one of the most important factors in maintaining customer satisfaction and loyalty. Social media, such as platform X, is often used by the public to convey their opinions and experiences regarding the services provided by PT Bank Rakyat Indonesia (Persero) Tbk. This research aims to analyze public sentiment towards the service performance of PT Bank Rakyat Indonesia (Persero) Tbk using the Support Vector Machine (SVM) algorithm. This research will also compare with the Multilayer Perceptron (MLP) algorithm. From the analysis of 2071 tweet data used, with the results of 838 neutral sentiments, 754 negative sentiments, and 479 positive sentiments. This shows that public perception of the performance of PT Bank Rakyat Indonesia (Persero) Tbk services tends to be neutral, with quite a lot of criticism compared to positive appreciation. This research uses TF-IDF word weighting. The training and testing process is carried out by dividing the training data and test data using a ratio of 80:20, 70:30, and 60:40. The best results were obtained from the SVM algorithm at 80:20 data division with an accuracy of 96.14%. Meanwhile, MLP managed to achieve an accuracy of 91.81% at the same ratio.

Actions (login required)

View Item