• This review investigates how machine learning (ML) technologies enhance financial transaction systems by automating risk analysis, fraud detection, credit evaluation, and algorithmic trading—topics central to digital transformation in financial information systems. • The study evaluates core ML techniques such as Support Vector Machines, Artificial Neural Networks, Genetic Algorithms, and Reinforcement Learning in terms of predictive accuracy, scalability, and organizational usability. • Through real-world use cases, the paper illustrates how ML models enable strategic decision-making, optimize operational efficiency, and facilitate data-driven insights in financial institutions. • Ethical and explainable AI (XAI) frameworks are critically examined to support trustworthy automation in compliance-sensitive environments. The paper explores how transparency and accountability can be embedded into AI-driven financial information systems. • The review identifies key research gaps in ML integration, including long-term forecasting, domain transferability, and interpretability—offering future research directions for scholars working on intelligent systems in financial and managerial contexts. Machine learning is transforming the financial sector by enhancing the decision-making ability, risk management, and operational efficiency. This study offers an in-depth analysis of the machine learning applications in the financial transactions, including fraud detection, credit scoring, algorithmic trading, financial forecasting, and regulatory compliance. Based on a comprehensive review of literature, this study offers methodologies, strengths, weaknesses, and future research directions. Industry case studies and applications reveal the significance of machine learning for finance. Further, the study underscores the significance of ethical business practices and explainable artificial intelligence (XAI) in financial system design as a basis of reliable financial systems. We explore machine learning for finance based on existing literature. The most frequently used methodologies in financial transaction research are Support Vector Machines (SVM), Artificial Neural Networks (ANN), and Genetic Algorithms (GAs), based on our evidence. We review the accomplishments and limitations of the existing literature. The conclusion of the survey identifies the gaps that currently exist and provides some suggestions for future research.
Sai et al. (Sun,) studied this question.