This study presents a comparative analysis of classical machine learning (ML) and quantum machine learning (QML) algorithms for industrial equipment fault detection using the AI4I 2020 Predictive Maintenance dataset. A total of nine models were evaluated through 5-fold stratified cross-validation: three classical models - Logistic Regression, Random Forest, and XGBoost - alongside six quantum models, namely Quantum Support Vector Machine (QSVM), Quantum K-Nearest Neighbors (QKNN), Quantum Kernel-based hybrid classifiers (QK+RF, QK+GB), Pegasos Quantum Support Vector Classifier (PegasosQSVC), and Variational Quantum Classifier (VQC). All quantum experiments were conducted using the IBM Qiskit framework and AerSimulator backend. Under a controlled comparison using an identical 300-sample budget and the same 5-fold cross-validation protocol, the best classical model (Random Forest, F1=0.902) outperformed the best quantum kernel method (QKNN, F1=0.820) by approximately 8.2 points. These results indicate that quantum kernel methods do not provide a dataefficiency advantage over classical methods on this dataset, although they achieve reasonably competitive results (F1 $\geq 0.75$ for kernel-based methods) at substantially higher computational cost (quantum: ~124-137 s/fold for kernel methods, up to ~257 s/fold for VQC, vs. classical: 0.01-0.26 s/fold). We analyze the sources of this gap and attribute it primarily to a mismatch between the unitarily-invariant geometry of the fidelity kernel and the axisaligned threshold structure of the AI4I failure modes, rather than to information loss in dimensionality reduction, which the faircomparison design rules out. We further show that this apparent quantum competitiveness stems partly from an unequal-data-volume comparison common in prior QML benchmarking studies, which our fair-comparison protocol corrects.
This study presents a comparative analysis of classical machine learning (ML) and quantum machine learning (QML) algorithms for industrial equipment fault detection using the AI4I 2020 Predictive Maintenance dataset. A total of nine models were evaluated through 5-fold stratified cross-validation: three classical models - Logistic Regression, Random Forest, and XGBoost - alongside six quantum models, namely Quantum Support Vector Machine (QSVM), Quantum K-Nearest Neighbors (QKNN), Quantum Kernel-based hybrid classifiers (QK+RF, QK+GB), Pegasos Quantum Support Vector Classifier (PegasosQSVC), and Variational Quantum Classifier (VQC). All quantum experiments were conducted using the IBM Qiskit framework and AerSimulator backend. Under a controlled comparison using an identical 300-sample budget and the same 5-fold cross-validation protocol, the best classical model (Random Forest, F1=0.902) outperformed the best quantum kernel method (QKNN, F1=0.820) by approximately 8.2 points. These results indicate that quantum kernel methods do not provide a dataefficiency advantage over classical methods on this dataset, although they achieve reasonably competitive results (F1 ≥0.75 for kernel-based methods) at substantially higher computational cost (quantum: ~124-137 s/fold for kernel methods, up to ~257 s/fold for VQC, vs. classical: 0.01-0.26 s/fold). We analyze the sources of this gap and attribute it primarily to a mismatch between the unitarily-invariant geometry of the fidelity kernel and the axisaligned threshold structure of the AI4I failure modes, rather than to information loss in dimensionality reduction, which the faircomparison design rules out. We further show that this apparent quantum competitiveness stems partly from an unequal-data-volume comparison common in prior QML benchmarking studies, which our fair-comparison protocol corrects.
This study systematically compares classical machine learning (ML) and quantum machine learning (QML) methods in the context of fault detection across all subsets of the NASA Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) dataset (FD001–FD004). Within a binary classification framework (RUL ≤ 30 cycles threshold), classical models such as Random Forest (RF), SMOTE+RF, Gradient Boosting (GBM), XGBoost, and LSTM are compared against quantum/hybrid models such as the Variational Quantum Classifier (VQC), Quantum Support Vector Machine (QSVM), Quantum LSTM (QLSTM), and Hybrid Quantum-Classical CNN (HQCNN). Experiments conducted using 4 qubits, 3 circuit layers, and a Noisy Intermediate-Scale Quantum (NISQ) simulator (PennyLane default.qubit) reveal that classical ML models—particularly XGBoost (F1 ≥ 0.940, AUC ≥ 0.990)—substantially outperform QML models in the current NISQ era (QLSTM: F1 ≈ 0.903, HQCNN: F1 ≈ 0.877). The ablation study demonstrates that this gap is attributable primarily to model architecture rather than to data volume. The study experimentally characterizes the current limitations of NISQ-era QML applications and offers practical guidance for the design of future hybrid pipelines.
A potential path forward is Quantum Machine Learning (QML), which aims to leverage quantum computing in conjunction with classical machine learning to enhance computing efficiency and the expressiveness of models. In this paper, two different quantum classifiers - Variational Quantum Classifier (VQC) and Quantum Kernel Support Vector Machine (QSVM) - are compared with three classical classifiers as baseline classifiers - Logistic Regression, Support Vector Machine (SVM), and a Multi-Layer Perceptron (MLP) - on the Breast Cancer Wisconsin dataset. The quantum circuits were created in the PennyLane framework and simulated on a classical backend. However, in terms of accuracy, classical Logistic Regression performed better with an accuracy of 97.8%, classical SVM and QSVM with an accuracy of 95.6% each, although the Quantum VQC achieved a lower accuracy of 88.9% and had a recall of 100% for the benign class, though it correctly identified only 12 of the 17 malignant cases (a malignant-class recall of approximately 70.6%). The drawback of quantum models is the higher training time; however, since the quantum circuit needs to be classically simulated, the quantum SVM took 23.29 seconds compared to less than 0.01 seconds for the classical linear models. These results indicate that for small structured datasets, classifiers based on quantum computing have not yet surpassed well-tuned classical counterparts. In some respects (e.g., benign-class recall), they perform competitively, though not on malignant-class recall, where the VQC in particular performed worse than the classical baselines, which is worth further investigation on real quantum computers.
Quantum machine learning has emerged as a promising paradigm that leverages quantum computational principles to potentially outperform classical approaches on specific learning tasks. This paper presents a comprehensive empirical investigation comparing Quantum Machine Learning (QML), specifically Quantum Support Vector Machines (QSVM) against Classical Machine Learning (CML) baselines including Logistic Regression, K Nearest Neighbors and classical SVMs. We implement quantum kernel methods using variational quantum circuits with angle embedding, IQP and ZZ feature maps on datasets exhibiting varying structural complexity. Our experiments on synthetic quantum generated datasets (Quantum Parity, Quantum Entanglement, Quantum Interference) demonstrate that QSVM achieves superior classification accuracy with improvements ranging from 2.9 % to 60 % over the best classical baselines achieving a 55.6 % overall win rate. Notably, on the Quantum Parity dataset, QSVM consistently outperformed all classical models across multiple configurations. On real-world benchmark datasets (Breast Cancer, Iris, Wine), QSVM Angle achieved 88% accuracy on Breast Cancer compared to 76% for classical SVM Linear a 15.8 % relative improvement while matching the best classical performance on Iris(100%)and Wine(92%). All experiments were conducted using quantum circuit simulation on classical hardware. These results provide empirical evidence that quantum kernel methods offer meaningful advantages over classical approaches on datasets with complex feature interactions. We discuss the theoretical foundations of these improvements including quantum feature space expressivity and kernel alignment.
Parkinson’s disease (PD) is a progressive neurodegenerative disorder affecting millions worldwide, where early and accurate diagnosis plays a critical role in improving patient outcomes. While classical machine learning methods have shown strong performance in PD classification using clinical and acoustic data, they remain limited in capturing complex nonlinear feature interactions in high-dimensional spaces. This study explores the application of quantum machine learning (QML) techniques for PD prediction, focusing on Variational Quantum Classifiers (VQC) and Quantum Support Vector Machines (QSVM), and compares their performance with strong classical baselines including Logistic Regression, SVM, Random Forest, Grid Search–optimized models, and XGBoost. Experiments were conducted using stratified train–test splits and validated through cross-validation and statistical testing. Results show that classical models achieve competitive performance, with XGBoost reaching 91.9% accuracy. However, quantum models demonstrate slightly improved performance, with the VQC using EfficientSU2 achieving the highest accuracy of 93.1%, followed by the Real Amplitudes VQC (92.6%) and QSVM (92.2%). These improvements are consistent across evaluation metrics including F1-score and recall. Statistical analysis using 10-fold cross-validation indicates that the VQC (EfficientSU2) shows a significant improvement over classical baselines ( p = 0.031), while other quantum models do not achieve statistical significance over XGBoost ( p > 0.05). McNemar’s test on the independent test set confirms no significant difference in final prediction distributions between quantum and classical models. Overall, the findings suggest that quantum machine learning models are competitive with state-of-the-art classical approaches, offering marginal but consistent performance gains. This study highlights the potential of quantum-enhanced learning frameworks for biomedical classification tasks, while also emphasizing that their advantages remain incremental rather than definitive under current simulation-based implementations.
Abstract This study presents a comparison between classical machine learning (ML) algorithms and their quantum-enhanced counterparts in classifying scikit’s breast cancer dataset. Specifically, the research focused on evaluating the accuracy and time computational efficiency of quantum-enhanced machine learning (QEML) algorithms against traditional or classical machine learning algorithms. The quantum algorithms employed include the quantum support vector machine (qSVM), variational quantum classifier (VQC) and quantum k-nearest neighbor (qkNN). The qSVM and qkNN algorithms discussed here are really kernel-based hybrid classical-quantum algorithms as opposed to fully quantum algorithms and the VQC is a hybrid classical-quantum algorithm. These algorithms were tested against their classical equivalents: support vector machine (SVM), k-nearest neighbor (kNN), and artificial neural network (ANN). The qSVM algorithm demonstrated an 90% accuracy in distinguishing malignant from benign cancer cases (the classical support vector machine algorithm had a 99% accuracy), while qkNN achieved 90% accuracy (the classical kNN algorithm had an accuracy of 97%), and VQC reported 88.77± 0.26% (averaged over 5 runs) (this was compared with an ANN which had an accuracy of 96%). The accuracies of the classical SVM, kNN, and ANN algorithms were recorded, to provide a basis for comparison. The results indicated that while QEML algorithms perform reasonably well in classification tasks, however they do not consistently outperform classical algorithms. Additionally, the study measured the time efficiency of each algorithm. The ANN required 33.8 seconds, kNN required 0.21 seconds, and SVM required 1.98 seconds to perform the classification tasks. In contrast, the quantum algorithms were significantly slower, with the VQC taking 632.53 seconds, qkNN requiring 495.34 seconds, and qSVM requiring 538.05 seconds (This is not fixed per run but give an impression of the time scales to run the algorithms). We emphasize that this simulation runtime does not reflect actual quantum hardware performance, where gate execution and circuit depth constraints differ significantly. To mitigate this ambiguity, we distinguish between simulation runtime and estimated quantum runtime where relevant. This work potentially highlights the cases where it may not offer any advantage to use quantum machine learning algorithms. The Receiver Operating Characteristic (ROC) curve and the Area Under the Curve (AUC) analysis are also considered in this work where the values measured for the SVM, kNN and ANN were 99%, 99% and 99% respectively and for the qSVM, qkNN and VQC were 87%, 86% and 86± 0.2% (averaged over 5 runs) respectively. Overall, this research contributes valuable insights into the comparative performance of classical and quantum ML algorithms in a specific application, underscoring the potential and current limitations of quantum approaches in machine learning tasks. Hyperparameter optimization was done for all the algorithms, both quantum and classical, with Optuna. This was primarily because the QEML algorithms had more parameters and required considerable computational resources to optimize. This could have influenced the performance outcomes and is the basis for future work. The development of key comparison criteria is critical in the development of quantum machine learning, and this work offers some building blocks for this.
Diabetes remains one of the common chronic diseases around the world and finding a good early prediction from clinical and lifestyle data is very important for timely help. Classical machine learning models such as regression, decision trees and random forests have been used for this purpose but the complicated non‑linear connections among clinical features might need richer ways of representation. Quantum Machine Learning (QML) gives an approach using quantum feature encoding and variational circuits to find patterns that are expensive to express with classical methods. In this study we show a diabetes prediction pipeline that compares baseline models—logistic regression, decision tree and random forest—with two quantum classifiers built in Qiskit: a Quantum Support Vector Classifier (QSVC) and a Variational Quantum Classifier (VQC). Using the Pima Indians Diabetes dataset we test all models on accuracy, precision, recall and F1‑score. The results show that the classical Random Forest classifier reaches the accuracy (0.73) beating both quantum classifiers. The Quantum Support Vector Classifier (QSVC 0.66) and the Variational Quantum Classifier (VQC 0.65) are still competitive. Do not outdo the classical baselines. These results agree with reports that in the present Noisy Intermediate‑Scale Quantum (NISQ) era quantum models usually match rather than surpass well‑tuned classical approaches on standard health‑care data and that a real quantum advantage will need larger data, deeper circuits and testing, on real quantum hardware.