Pandey Trilok Nath, Ravalekar Vishvajeet, Nair Sidharth D, Pradhan Sunil Kumar
School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, Tamilnadu, 600127, India.
School of Electronics Engineering, Vellore Institute of Technology, Chennai, Tamilnadu, 600127, India.
Sci Rep. 2025 Aug 4;15(1):28443. doi: 10.1038/s41598-025-12515-4.
This research paper delves into the realm of quantum machine learning (QML) by conducting a comprehensive study on time-series data. The primary objective is to compare the results and time complexity of classical machine learning algorithms on traditional hardware to their quantum counterparts on quantum computers. As the amount and complexity of time-series data in numerous fields continues to expand, the investigation of advanced computational models becomes critical for efficient analysis and prediction. We employ a time-series dataset that include temperature records from different nations throughout the world spanning the previous half of the century. The study compares the performance of classical machine learning algorithms to quantum algorithms, which use the concepts of superposition and entanglement to handle subtle temporal patterns in time-series data. This study attempts to reveal the different benefits and drawbacks of quantum machine learning in the time-series domain through rigorous empirical analysis. The findings of this study not only help to comprehend the applicability of quantum algorithms in real-world contexts, but they also open the way for future advances in utilizing quantum computing for increased time-series analysis and prediction. This study's findings could have ramifications in industries ranging from finance to healthcare, where precise forecasting using time-series data is critical for informed decision-making.
本研究论文通过对时间序列数据进行全面研究，深入探讨了量子机器学习（QML）领域。主要目标是比较传统硬件上经典机器学习算法与量子计算机上量子对应算法的结果和时间复杂度。随着众多领域中时间序列数据的数量和复杂性不断增加，对先进计算模型的研究对于高效分析和预测变得至关重要。我们使用了一个时间序列数据集，其中包括上个世纪前半叶来自世界不同国家的温度记录。该研究将经典机器学习算法的性能与量子算法进行了比较，量子算法利用叠加和纠缠的概念来处理时间序列数据中的微妙时间模式。本研究试图通过严格的实证分析揭示量子机器学习在时间序列领域的不同优缺点。这项研究的结果不仅有助于理解量子算法在现实世界中的适用性，还为未来利用量子计算进行增强的时间序列分析和预测开辟了道路。这项研究的结果可能会对从金融到医疗保健等行业产生影响，在这些行业中，使用时间序列数据进行精确预测对于明智的决策至关重要。
