Artificial Intelligence in Transactional Data Analysis: A Data-Centric Analysis of Customer Behavior in the U.S.A.

Abstract

The rapid expansion of digitalization in the financial services sector has changed the relationship way of customers with banks while creating new opportunities and challenges regarding the understanding of transactional behavior. This work will explore Artificial Intelligence in Transactional Data Analysis: A Data-Centric Analysis of Customer Behavior using composite models that represent demographic, temporal, and geographic context along with predictive modeling. The study utilizes descriptive statistics, ANOVA, clustering unity, and time-series forecasting on data from U.S. bank customers on Kaggle, contributing clarification to multiple dimensions with more than 2,500 observations supplemented by Federal Deposit Insurance Corporation (FDIC) and Federal Reserve data. The life-cycle patterns are evident that the middle-aged have classified as the highest balances and volumes, whereas transactions with students are numerous but lower in value and seniors are still heavily branch dependent. Temporal patterns reveal daily transactions that take place between Tuesday and Thursday are higher than other days of the week and the seasonal peaks in August-September. While urban centers have demonstrated the most successful adoption and integration of digital technologies, rural populations remain excluded. Predictive models estimate a churn risk as around 20% and indicate that the problem is associated with the inactive customers, whereas clustering analysis isolates a pattern of four different behavioral groups of customers that have their own managerial relevance. The study uses smart ideas about the Life-Cycle Hypothesis, Technology Acceptance Model (TAM) and Behavioral Economics to show how AI can be used to analyze banking transactions. It also focuses on making sure the AI is ethical that it is fair, easy to understand, and works for everyone. This makes the research useful both for scholars and for practical use in the real world. These results provide banks with data-driven metrics by which to improve retention, improve the investment between digital and physical channels, and incorporate best practices for the deployment of responsible AI that builds customer trust and helps financial inclusion.