European Journal of Computer Science and Information Technology (EJCSIT)

EA Journals

feature engineering

Fraud Detection in Financial Services Using Advanced Machine Learning (Published)

Fraud detection in financial services has evolved substantially with the integration of advanced machine learning techniques, replacing traditional rule-based systems that have shown diminishing effectiveness in recent years. This transformation has been driven by the exponential growth in transaction volume, velocity, and variety across digital financial ecosystems. Machine learning models, particularly ensemble techniques like Isolation Forests and XGBoost, alongside deep learning architectures such as autoencoders and neural networks, have demonstrated remarkable capabilities in identifying fraudulent patterns while significantly reducing false positives. The article examines how sophisticated feature engineering processes, including transaction velocity tracking, merchant category analysis, and device fingerprinting, serve as critical foundations for effective fraud detection. The challenges of extreme class imbalance are addressed through innovative resampling techniques and cost-sensitive learning frameworks. Operational implementation considerations, including real-time processing constraints, multi-layered architecture design, and the emerging role of graph-based fraud network analysis, are explored in depth. The findings reveal that optimized machine learning approaches not only enhance fraud detection rates but also minimize customer friction while meeting strict regulatory requirements for model explainability.

Keywords: class imbalance, feature engineering, financial fraud detection, graph-based network analysis, machine learning ensemble models, real-time decision systems

A Comprehensive Guide to Optimizing Machine Learning and Deep Learning Models (Published)

Machine learning and deep learning model optimization remain a pivotal aspect of artificial intelligence development, encompassing crucial elements from data preprocessing to deployment monitoring. The optimization process involves multiple interconnected stages, including data quality management, algorithm selection, feature engineering, hyperparameter tuning, transfer learning, and model deployment strategies. Each stage presents unique challenges and opportunities for enhancing model performance, with modern techniques offering solutions for improved accuracy, efficiency, and reliability. From addressing data quality issues through systematic preprocessing to implementing sophisticated deployment monitoring systems, the various aspects of model optimization work together to create robust and effective machine learning solutions that can be successfully deployed in real-world applications.

Keywords: MLOps deployment, feature engineering, hyperparameter tuning, model optimization, transfer learning

Scroll to Top

Don't miss any Call For Paper update from EA Journals

Fill up the form below and get notified everytime we call for new submissions for our journals.