International Symposium on Digital Forensics and Security (ISDFS)

Exploring the Potential of Quantum-Based Machine Learning: A Comparative Study of QSVM and Classical Machine Learning Algorithms

Zeynep Özpolat, Murat Karabatak

Published May 11, 2023• 9 citations

Abstract

Research using innovative, rapidly evolving quantum-based systems is becoming increasingly important. The acceleration provided by the real use of a few quantum computers in the world constitutes very important steps for the computer world. Quantum-based systems are also frequently used in data science and machine learning. In this study, Quantum Support Vector Machine (QSVM), a quantum-based machine learning algorithm, is compared with classical machine learning algorithms. IBM - Studies on quantum computer simulation have been carried out. The data dimension is reduced by first applying Principal Component Analysis (PCA) to the dataset. Afterwards, analyzes were made and the results were presented comparatively. The highest performance of the QSVM algorithm was obtained at 92.3%. Compared to classical machine learning algorithms, the performance of the QSVM algorithm is slightly lower. It is thought that higher performance and speed will be achieved in QSVM by eliminating limitations such as limited qubit resources and system restrictions. It is clear that the results found are promising for quantum-based machine learning algorithms.

DOI: 10.1109/isdfs58141.2023.10131821

Related papers

Quantum Machine Learning: Enhancing Algorithms through Quantum Neural Networks and Data Analysis

Rakesh Sengupta, Usha Desai, Anand Singh Rajawat +3 more

This paper evaluates the performance of two quantum machine learning algorithms: Quantum Support Vector Machine (QSVM) and Quantum Principal Component Analysis (Q-PCA). The scalability of QSVM is analyzed across increasing dataset sizes and feature dimensions, revealing faster-than-linear growth in kernel evaluation and SVM training times. However, this computational overhead is accompanied by a notable decline in classification accuracy, highlighting challenges in applying QSVM to larger or more complex datasets. In contrast, Q-PCA demonstrates near-linear scalability with dataset size and qubit count, leveraging the independence of quantum circuit execution and structured operations to maintain efficiency. A QNN-based implementation of Q-PCA further validates its ability to align data with principal components, maximize variance, and enable effective dimensionality reduction. These findings emphasize the potential of Q-PCA for efficient quantum-enhanced data analysis while underscoring the need for optimization in QSVM and further exploration of Q-PCA’s capabilities for real-world quantum computing applications.

International Conference on Augmented Reality, Intelligent Systems, and Industrial Automation (ARIIA)Dec 20, 20243 citations

Comparison of QSVM with Other Machine Learning Algorithms on EEG Signals

Gamzepelin Aksoy, Murat Karabatak

In the Brain-Computer Interfaces where Electroencephalogram (EEG) signals are used, fast and accurate analysis of data is one of the important steps. For this aim, various machine learning methods are being developed. In this study, we compared the performance of the Quantum Powered Support Vector Machine (QSVM) with classical machine learning algorithms for classifying EEG data. Analyzes were performed in the quantum simulation environment and due to the inadequacies in the quantum environment, the number of samples in the data set was reduced by using Principal Component Analysis. The maximum performance of QSVM was found to be 99.47% in the experiments. When all the results in the analyzes are examined, it is seen that the QSVM has achieved good performance. By removing the limitations of quantum computers and making changes in the hyperparameters of quantum-based algorithms, the use of quantum-based machine learning methods in the classification of EEG data can provide better results in terms of accuracy and runtime.

International Symposium on Digital Forensics and Security (ISDFS)May 11, 202310 citations

A review of quantum support vector machine for data classification

Tran, Tien

Quantum Machine Learning is an active research field where the power of quantum mechanics is integrated into machine learning algorithms to solve problems that are intractable for classical computers. Quantum support vector machine (QSVM) was one of the first quantum algorithms that got implemented on real quantum hardware. It demonstrated promising outcomes as compared to classical Support vector machine (SVM) through many implementations. However, it encounters challenges in scaling with large datasets and in resource effectiveness when compare to the usage of innovative classical algorithms in the Noisy Intermediate Scale Quantum era. Here is a review of some latest developments of QSVM and discussion on prospects of QSVM application in the future.

Aug 30, 20240 citations

Design and Implementation of Efficient Quantum Support Vector Machine

Mhlambululi Mafu, Makhamisa Senekane

Machine Learning (ML) is arguably the most ad-vanced sub-field of Artificial Intelligence (AI). It concerns the study of computer algorithms that can improve automatically by learning from data (experience) without being explicitly pre-programmed. On the other hand, the quantum version of ML, Quantum Machine Learning (QML), forms one of the most crucial and recent quantum computing applications. It uses quantum mechanical principles such as superposition, interference, and tunneling to enable quantum computers to learn from data. Some examples of QML algorithms include Quantum Principal Component Analysis (QPCA) and Quantum Support Vector Machine (QSVM). The QSVM is a robust supervised machine learning algorithm used for classification and regression. Therefore, this paper discusses the design and implementation of an efficient QSVM model. The model's efficiency is achieved using Principal Component Analysis (PCA) on the classical data, then loading the classical data into a quantum computer. Experimental results demonstrate the significance and application of the QSVM model reported in this paper.

International Conference on Electrical, Computer and Energy Technologies (ICECET)Dec 9, 202125 citations

Quantum-Based Feature Extraction for Machine Learning

Jincheng Zhang

This paper investigates the potential of leveraging quantum algorithms for feature extraction in machine learning. Traditional feature extraction methods often struggle with high-dimensional data and computational complexity, leading to performance bottlenecks in downstream machine learning tasks. We explore the application of quantum algorithms, specifically Quantum Principal Component Analysis (QPCA) and Quantum Support Vector Machines (QSVM), to extract more efficient and potentially more informative features from datasets. The core claim of this research is that utilizing quantum algorithms can significantly enhance the performance of machine learning models. The mechanism involves applying quantum algorithms to reduce the dimensionality of data while preserving crucial information, which is then fed into classical machine learning models. This work provides a foundational exploration into this emerging field, demonstrating the potential for increased efficiency and accuracy in machine learning applications. ---

Sep 3, 20260 citations

Comparative Study of Classical and Quantum Machine Learning Models: Insights into Quantum Advantage in Materials Informatics

Aris Tri Jaka Harjanta, Hindriyanto Dwi Purnomo, Hendry

Quantum Machine Learning (QML) has emerged as a promising paradigm for addressing increasing computational and representational demands in materials informatics. While classical models such as Support Vector Machines (SVM) achieve strong predictive performance, they often struggle to capture complex, highly correlated interactions in high-dimensional materials data. QML addresses this challenge by leveraging quantum-mechanical principles to construct expressive feature embeddings, where prospective quantum advantage lies in generating feature spaces that are difficult to approximate classically. In this study, 1,000 crystalline compounds from the Open Quantum Materials Database (OQMD) are evaluated in a binary classification task based on formation-energy stability. The dataset is normalized, reduced to four dimensions via Principal Component Analysis (PCA), and encoded into quantum circuits. Three QML models—QSVM, VQC, and QNN—are benchmarked against a classical SVM using repeated stratified evaluation. Results show that the classical SVM achieves the highest accuracy (91.8% ± 0.012), followed by QSVM (60.8% ± 0.035), while VQC and QNN perform significantly worse. This gap is driven by limited qubit capacity, encoding inefficiencies, restricted circuit expressivity, and optimization challenges. Nevertheless, QSVM demonstrates stable performance, suggesting that potential quantum advantage may emerge from improved feature encoding and kernel design rather than deeper variational circuits.

Advance Sustainable Science Engineering and TechnologyApr 30, 20260 citations