The problem of imbalanced datasets is a significant concern when creating reliable credit card fraud (CCF) detection systems. In this work, we study and evaluate recent advances in machine learning (ML) algorithms and deep reinforcement learning (DRL) used for CCF detection systems, including fraud and non-fraud labels. Based on two resampling approaches, SMOTE and ADASYN are used to resample the imbalanced CCF dataset. ML algorithms are, then, applied to this balanced dataset to establish CCF detection systems. Next, DRL is employed to create detection systems based on the imbalanced CCF dataset. The diverse classification metrics are indicated to thoroughly evaluate the performance of these ML and DRL models. Through empirical experiments, we identify the reliable degree of ML models based on two resampling approaches and DRL models for CCF detection. When SMOTE and ADASYN are used to resampling original CCF datasets before training/test split, the ML models show very high outcomes of above 99% accuracy. However, when these techniques are employed to resample for only the training CCF datasets, these ML models show lower results, particularly in terms of logistic regression with 1.81% precision and 3.55% F1 score for using ADASYN. Our work reveals the DRL model is ineffective and achieves low performance, with only 34.8% accuracy.

In the modern business environment, artificial intelligence (AI) and machine learning (ML) are widely applied in tasks in customer behavior analysis (CBA), such as purchase intention, churn prediction, customer segmentation and profiling, and customer satisfaction analysis. However, despite robust predictive effectiveness, many AI/ML-based CBA systems remain model-centric and stop at predictive outputs, leaving a critical gap between analytical predictions and managerial decision-making. To address this gap, this study offers a systematic and structured literature review of AI/ML-driven CBA research based on PRISMA. First, this study redefines AI/ML-based CBA as a form of decision support rather than isolated prediction. This study synthesizes existing research by explicitly linking predictive tasks to the managerial decision-making context, thereby clarifying the persistent gap between prediction and decision. Secondly, the study develops an architecture for AI/ML-oriented expert systems and intelligent decision support systems. This architecture highlights business rules, organizational constraints, human-in-the-loop validation, and explainable AI, where explainability is indicated as a core decision enabler rather than a post-hoc analytical tool. Third, the research proposes a decision-aware evaluative framework. Beyond accuracy metrics, model evaluation also examines decision effectiveness, governance requirements, and organizational impact. This study also indicates major challenges and future research directions for CBA involving decision support systems.