Financial fraud detection remains difficult due to extreme class imbalance and subtle, nonlinear patterns in transactional data. We investigate a hybrid quantum–classical approach that couples a fidelity-based quantum kernel with a logistic regression classifier (QEKLR). A custom ZZ-inspired feature map comprising Hadamards, mixed-axis Ry/Rzencodings, tunable Ising-ZZ cross terms, and data re-uploading, embeds PCA-reduced, Min–Max scaled features into quantum states. The resulting fidelity kernel is evaluated and used within kernel logistic regression. Experiments on two benchmarks: BankSim (synthetic) and the European Credit Card Fraud dataset, use balanced subsets created via Random Under-Sampling. The proposed model attains the best performance on BankSim (AUC = 0.96; strongest F1), and is competitive on the credit-card dataset (AUC = 0.94), closely tracking XGBoost (AUC = 0.96) while outperforming logistic regression and a variational quantum classifier baseline. These results indicate that quantum kernel methods can provide practical gains on tabular fraud tasks while remaining aligned with current NISQ constraints. We also discuss the quadratic cost of exact kernel evaluation and outline scalable approximations for future deployment on larger, highly imbalanced datasets.
Financial fraud detection remains difficult due to extreme class imbalance and subtle, nonlinear patterns in transactional data. We investigate a hybrid quantum–classical approach that couples a fidelity-based quantum kernel with a logistic regression classifier (QEKLR). A custom ZZ-inspired feature map comprising Hadamards, mixed-axis Ry/Rzencodings, tunable Ising-ZZ cross terms, and data re-uploading, embeds PCA-reduced, Min–Max scaled features into quantum states. The resulting fidelity kernel is evaluated and used within kernel logistic regression. Experiments on two benchmarks: BankSim (synthetic) and the European Credit Card Fraud dataset, use balanced subsets created via Random Under-Sampling. The proposed model attains the best performance on BankSim (AUC = 0.96; strongest F1), and is competitive on the credit-card dataset (AUC = 0.94), closely tracking XGBoost (AUC = 0.96) while outperforming logistic regression and a variational quantum classifier baseline. These results indicate that quantum kernel methods can provide practical gains on tabular fraud tasks while remaining aligned with current NISQ constraints. We also discuss the quadratic cost of exact kernel evaluation and outline scalable approximations for future deployment on larger, highly imbalanced datasets.
The increasing dependence on digital and card-based payment systems has heightened the risk of financial fraud. This situation creates an urgent need for models that can identify rare fraudulent events in uneven transaction data. Traditional machine learning models often sacrifice recall for precision when the minority class is very small. Quantum classifiers are theoretically appealing for capturing complex feature relationships, but current hardware limitations on qubit count and circuit depth restrict their effectiveness. This work introduces a hybrid quantum-classical framework for credit card fraud detection. It combines an Extreme Gradient Boosting (XGBoost) classifier with a Quantum Support Vector Machine (QSVM) that uses a ZZ feature map and a fidelity-based quantum kernel. To handle class imbalance, we apply the Synthetic Minority Oversampling Technique (SMOTE) and adjust decision thresholds. The quantum element is trained on a reduced, class-balanced data subset, ensuring it stays within the feasible limits of the simulator. We merge predictions from the classical and quantum models using a weighted probability fusion approach, optimizing the final threshold to maximize the F1-score. Using the Kaggle Credit Card Fraud Detection dataset, the hybrid model achieves a precision of 0.9783, recall of 0.9000, F1-score of 0.9375, and ROC-AUC of 0.9805. It surpasses the standalone XGBoost baseline, XGBoost with SMOTE, and the standalone QSVM regarding F1 and ROC-AUC metrics. These results show that blending classical gradient boosting with quantum kernel-based classification, along with imbalance-aware techniques, can significantly enhance fraud detection quality despite current quantum hardware limitations. This offers a practical path for future work in quantum-assisted financial security.
Detecting credit card fraud is notoriously difficult because fraudulent charges make up less than 0.2% of total transactions. Quantum Machine Learning (QML), particularly Quantum Kernel Support Vector Machines (QSVM), promises richer feature representations by embedding classical data into high-dimensional Hilbert spaces. However, whether this theoretical advantage holds up when managing real-world financial data bottlenecks remains unclear. We designed a controlled benchmark comparing three quantum encodings, Angle, ZZFeatureMap, and IQP against classical models while keeping sample sizes, feature counts, and scaling strictly matched. To handle class skew without corrupting quantum state overlaps, we embedded synthetic minority oversampling (SMOTE) directly into the quantum Gram matrix workflow, raising PR-AUC scores by roughly +0.04 across feature maps. When trained on the full European credit card dataset with 30 features, classical baselines reached a top PR-AUC of 0.983. However, restricting classical models to match QSVM hardware constraints (500 samples, 4 PCA features) dropped their PR-AUC to 0.750. This puts our best quantum result 0.730 PR-AUC using IQP encoding on 150 samples into realistic context. Testing on physical quantum hardware reduced performance further to 0.62 due to CNOT gate errors and thermal noise. Overall, our findings pinpoint two main operational bottlenecks for near-term quantum kernels: the scaling of Gram matrix evaluation and heavy information loss from squeezing features onto small qubit registers.
Credit card fraud detection is still one of the significant challenges in the digital economy, and most traditional machine learning methods appear to be inadequate fordealing with highly imbalanced datasets and intricate non-linear fraud patterns. This paper introduces and tests Q-fusion enhanced hybrid models, which exploit Quantum-Inspired FeatureMaps to embed classical transaction data in high-dimensional Hilbert spaces with the aim of increasing class distinguishability and performance. Five quantum-boosting classifiers were performed includingand described: Quantum Support Vector Machine (QSVM), Quantum Logistic Regression (QLR), Quantum K-Nearest Neighbors (QKNN), Quantum Gaussian Naive Bayes (QGNB) and Quantum Multi-Layer Perceptron (QMLP). The results show that the Quantum MLP model provided the best performance and reached an accuracy of 99.35%, precision of 0.96, recall of0.92, F1-score of 0.94 and ROC-AUC score of 0.97 which outperformed all quantum-enhanced models as well as classical ANN model.. Quantum feature mapping greatly improved feature separability, especially for minority class fraudulent transactionsand remained compatible with classical infrastructure. These results identify quantum-inspired machine learning as a promising route for enhancing fraud detection in severely imbalanced data sets and can berealized without deploying full-scale quantum hardware. implementation faces hardware limitations ( 27), algorithmic complexity (28), and infrastructure integration challenges (29). Quantum-inspired approaches provide a pragmatic pathway, offering quantum-like advantages through classical simulation (30; 31). Research Problem and Contribution Credit card fraud detection faces interconnected challenges: (1) extreme class imbalance creates severe classifier bias where models achieve high accuracy by predicting all transactions as legitimate while missing fraud (32; 33); (2) feature engineering requires domain expertise to identify transaction attributes, temporal patterns, and behavioral anomalies (14; 34); (3) real-world systems require millisecond classification (35); (4) regulatory compliance demands transparent, explainable systems (36; 37). Despite growing QML interest, empirical evaluations for fraud detection remain limited. Most studies focus either on purely classical methods (8; 32) or theoretical quantum algorithms requiring full hardware (30; 38), creating a gap in practical quantum-inspired implementations. Recent surveys identify this as a critical frontier (26; 39; 40). This research contributions: (1) Implementation and evaluation of five quantum-enhanced hybrid models (QSVM, QLR, QKNN, QGNB, QMLP); (2) Demonstration that quantum-inspired transformations project data into higher-dimensional Hilbert spaces, improving fraud class separability; (3) Empirical evidence of substantial gains (99.35% accuracy, 0.97 ROC-AUC) on classical hardware; (4) Rigorous comparative evaluation with consistent preprocessing and comprehensive metrics; (5) Identification that Quantum MLP achieves synergistic performance combining quantum features with deep learning. Research Objectives This research aims to: (1) develop quantum-inspired feature mapping transforming transaction data into Hilbert space representations; (2) design and evaluate five quantum-enhanced classifiers; (3) conduct comprehensive performance evaluation using imbalanced classification metrics; (4) quantify performance improvements from quantum-inspired transformations; (5) assess computational feasibility for production deployment; (6) provide theoretical explanations based on Hilbert space geometry. Paper Organization Section 2 reviews related work in machine learning fraud detection, quantum computing, and quantum machine learning. Section 3 describes methodology including preprocessing, quantum feature mapping, classifier architectures, and evaluation metrics. Section 4 presents experimental results with performance analysis. Section 5 concludes with key findings and future directions. 2 Related Work 2.1 Classical Machine Learning for Fraud Detection Credit card fraud detection has been extensively studied using classical approaches. Early work established cost-sensitive learning to address class imbalance (41). Logistic Regression provides interpretability and probabilistic outputs (3) but struggles with nonlinear patterns (7). Support Vector Machines identify maximum-margin hyperplanes (5) but face computational challenges (O(n 2 )-O(n 3 )) (42) and performance degradation on imbalanced data (8). Ensemble methods including Random Forest (4) and Gradient Boosting (43; 44) achieve improved stability. Deep learning approaches-Multi-Layer Perceptrons with modern optimization (ReLU (45), batch normalization (46), dropout (47), Adam (48))-achieve state-of-the-art performance with preprocessing (49). Recurrent architectures (LSTM (50), GRU (51)) capture temporal patterns (52), while Autoencoders frame fraud as anomaly detection (53; 54). Imbalance Handling Techniques Class imbalance (¡0.2% fraud) (13) necessitates specialized techniques. SMOTE generates synthetic minority samples through interpolation (6), while SMOTEENN combines SMOTE with Edited Nearest Neighbors for noise reduction (55; 56). Cost-sensitive learning assigns different misclassification penalties (33; 57). Ensemble methods train multiple classifiers on balanced subsets (58; 59).
This paper investigates whether hybrid quantum-classical machine learning can deliver practical improvements in financial fraud detection performance for card-based and other payment transactions. Building on a Guided Quantum Compressor architecture, the approach integrates an autoencoder, a variational quantum circuit, and a classical neural head, and then embeds this hybrid model into a mixture-of-experts framework including a state-of-the-art gradient-boosted tree classifier. Using a European credit card dataset with severe class imbalance, the routed hybrid architecture with 0.6 threshold achieves average precision scores of 0.793±0.085 compared to 0.770±0.096 of XGBoost on 3 repeated 5-fold cross-validation benchmarks. Precision and recall comparisons reveals a possible trade-off of fraud and nominal detections with a reduction in false positives at the cost of a small reduction in fraud detections. The improvements are achieved while adding only 7 to 21 minutes of extra inference time depending on the choice of hyperparameters. These results indicate that selectively routing transactions to quantum-classical models can enhance fraud detection while remaining compatible with the latency and operational constraints of modern financial institutions.
The increasing complexity and volume of financial transactions pose significant challenges to traditional fraud detection systems. This technical report investigates and compares the efficacy of classical, quantum, and quantum-hybrid machine learning models for the binary classification of fraudulent financial activities. As of our methodology, first, we develop a comprehensive behavioural feature engineering framework to transform raw transactional data into a rich, descriptive feature set. Second, we implement and evaluate a range of models on the IBM Anti-Money Laundering (AML) dataset. The classical baseline models include Logistic Regression, Decision Tree, Random Forest, and XGBoost. These are compared against three hybrid classic quantum algorithms architectures: a Quantum Support Vector Machine (QSVM), a Variational Quantum Classifier (VQC), and a Hybrid Quantum Neural Network (HQNN). Furthermore, we propose Fraud Detection for Quantum Computing (FD4QC), a practical, API-driven system architecture designed for real-world deployment, featuring a classical-first, quantum-enhanced philosophy with robust fallback mechanisms. Our results demonstrate that classical tree-based models, particularly \textit{Random Forest}, significantly outperform the quantum counterparts in the current setup, achieving high accuracy (97.34%) and F-measure (86.95%). Among the quantum models, \textbf{QSVM} shows the most promise, delivering high precision (77.15%) and a low false-positive rate (1.36%), albeit with lower recall and significant computational overhead. This report provides a benchmark for a real-world financial application, highlights the current limitations of quantum machine learning in this domain, and outlines promising directions for future research.
With the rapid growth of digital financial transactions, credit card fraud detection has become a critical challenge in the banking and finance sector. Traditional rule-based systems and classical machine learning (ML) models often struggle to detect complex and evolving fraud patterns, especially in highly imbalanced datasets. This paper proposes a novel hybrid approach to credit card fraud detection using Quantum Machine Learning (QML) techniques combined with classical methods. The system applies Principal Component Analysis (PCA) for dimensionality reduction and Synthetic Minority Oversampling Technique (SMOTE) to handle class imbalance. A classical Random Forest model is implemented as a baseline, while a QML model based on Variational Quantum Circuits (VQC) is developed using PennyLane. A hybrid decision engine combining predictions of both models is designed for improved performance. Experimental results demonstrate that the hybrid model achieves 96.9% accuracy, outperforming individual models by achieving better detection rates and reducing false predictions. The system is deployed via an interactive Streamlit-based dashboard. This work highlights the potential of Quantum Machine Learning in enhancing fraud detection systems in financial security applications.