• Mamta Indrasing Girase
• Jagruti Sarang Patil

The rapid growth of digital payment systems and online financial transactions has increased the exposure of financial institutions and customers to credit card fraud. Traditional rule-based fraud detection systems are often unable to adapt effectively to evolving fraudulent behavior and may generate a high number of false alarms. Machine learning (ML) provides a data-driven alternative by learning transaction patterns and identifying suspicious activities automatically. This paper presents an imbalance-aware machine learning framework for credit card fraud detection using Logistic Regression, Decision Tree, Random Forest, and Extreme Gradient Boosting (XGBoost). The study is based on the widely used Credit Card Fraud Detection dataset containing 284,807 transactions, of which 492 are fraudulent, resulting in an extreme class imbalance of approximately 0.172%. The dataset contains anonymized PCA-transformed variables V1–V28 along with Time, Amount, and the binary Class target variable. The proposed framework consists of data preprocessing, feature scaling, stratified train-test splitting, class-imbalance handling, model training, hyperparameter optimization, and evaluation using confusion matrix, precision, recall, F1-score, ROC-AUC, and Precision-Recall AUC. Particular emphasis is placed on precision and recall because conventional accuracy can be misleading for highly imbalanced fraud datasets. The framework also supports real-time deployment in which a transaction is processed, transformed into model-ready features, assigned a fraud probability, and forwarded to an appropriate decision layer. The study provides a systematic methodology for comparing classical and ensemble machine learning algorithms and establishes a foundation for developing adaptive, scalable, and interpretable fraud detection systems.

Credit Card Fraud Detection, Machine Learning, Class Imbalance, Random Forest, XGBoost, Logistic Regression, SMOTE, Fraud Detection, Precision, Recall, Explainable AI.

"“An Imbalance-Aware Machine Learning Framework for Credit Card Fraud Detection” ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.13, Issue 9, page no.a14-a24, September-2026, Available :http://www.jetir.org/papers/JETIR2609003.pdf

"“An Imbalance-Aware Machine Learning Framework for Credit Card Fraud Detection” ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.13, Issue 9, page no. ppa14-a24, September-2026, Available at : http://www.jetir.org/papers/JETIR2609003.pdf