Peningkatan kualitas layanan perbankan digital melalui pengelompokan tweet menggunakan DBSCAN

Abstract

Digitalization in the banking sector allows customers to obtain banking services independently without having to come directly to the bank. Digital banking services enable customers to obtain information, communicate, register, open accounts, banking transactions and close accounts, including obtaining other information and transactions outside of banking products. Banking is intensively providing services or promotions through social media, one of which is by using Twitter social media. Density-Based Spatial Clustering of Applications with Noise (DBSCAN) method. DBSCAN clustering is done by combining the Eps and MinPts values to produce the highest silhouette coefficient. The highest silhouette coefficient values from BRI, Mandiri and BCA banking service tweets produce 33, 14, and 39 clusters, respectively, with different Eps and MinPts values. Based on the results of wordcloud, it shows that banking services need to be improved in terms of checking DM on accounts, customers ask the admin to immediately respond to complaints related to ATM cards, disruptions to mobile banking and some say thank you for the services that have been provided.