Abstract
This study develops a composite Digital Banking Index for Vietnamese commercial banks using text mining and principal component analysis. The dataset comprises 330 annual reports from 30 banks during 2015 - 2025. Keyword density was calculated across four dimensions: technological infrastructure, digital channels and customer experience, digital operations, and digital strategy and human resources. Principal component analysis was then applied to generate an objective composite index without subjectively assigned weights. Validation results demonstrate satisfactory reliability and confirm that the index captures the multidimensional nature of banking digitalization. The findings also identify three development stages: basic digitalization during 2015 - 2018, rapid expansion during 2019 - 2022, and deeper but increasingly differentiated digital transformation during 2023 - 2025. The index provides a reliable dataset for subsequent econometric research on banking performance and systemic risk.
: banking digitalization
Digital Banking Index
text mining
principal component analysis
Vietnamese commercial banks