Details

Autor(en) / Beteiligte

• Alrizq, Mesfer
• Alghamdi, Abdullah

Titel

Customer satisfaction analysis with Saudi Arabia mobile banking apps: a hybrid approach using text mining and predictive learning techniques

Ist Teil von

• Neural computing & applications, 2024-04, Vol.36 (11), p.6005-6023

Ort / Verlag

London: Springer London

Erscheinungsjahr

2024

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• Mobile banking is becoming increasingly popular as a new means of delivering financial services, particularly in places where many people lack access to traditional banking institutions. In the current banking industry, mobile banking software, sometimes known as apps, has largely replaced traditional branch banking services. Consumers’ banking experiences can be greatly enhanced, and bank processes can be simplified, with the advent of mobile banking via apps. Customers’ satisfaction with mobile banking apps has been an important issue in recent research. In addition, the assessment of mobile apps in online banking has gained significant popularity. There have been several studies on customers’ satisfaction with online banking; however, this issue is not widely investigated by machine learning techniques. Specifically, there is no study to investigate customers’ satisfaction with mobile banking apps and evaluate them using a comprehensive set of factors using predictive text mining and machine learning techniques. In this study, we develop a hybrid method using text mining and regression machine learning approaches to evaluate the factors impacting customers’ satisfaction with online banking apps in Saudi Arabia. The factors are discovered from users generated content (UGC) in mobile banking apps using latent Dirichlet allocation (LDA). The customers’ satisfaction is predicted using support vector regression (SVR) and principal component analysis (PCA). The results show that machine learning can be an effective approach to assessing customers’ satisfaction with online banking apps using the factors discovered by text mining from UGC.