This paper presents a comprehensive study on the detection of malicious URLs using both classical and quantum machine learning (QML) techniques. The research focuses on identifying fraudulent URLs that mimic legitimate ones, which are commonly used in phishing attacks. The study evaluates various machine learning algorithms, including logistic regression, decision trees, support vector machines (SVM), and neural networks, to achieve high accuracy in detecting malicious URLs. The results show that these classical methods can achieve true positive rates greater than 90% when applied to different datasets. The study also explores the application of quantum machine learning for URL detection. Quantum machine learning offers a promising approach by leveraging quantum computing principles, such as superposition and entanglement, to potentially improve the efficiency and accuracy of classification models. The research introduces a Variational Quantum Classifier (VQC), which combines quantum circuits with classical optimization techniques to perform binary classification of URLs. The VQC was tested on a dataset of URLs, and the results showed that it could achieve high accuracy, although it was found that classical models generally outperformed quantum models in this specific task. The study highlights the importance of data preprocessing, including feature selection and encoding, to ensure the effectiveness of both classical and quantum models. The dataset used in the study contains information about URLs, including their length, content, and server type, and was preprocessed to remove irrelevant features and balance the dataset. The results indicate that the dataset is unbalanced, with a higher number of fraudulent URLs than non-fraudulent ones, which affects the performance of the models. The research concludes that while classical machine learning models are currently more effective in detecting malicious URLs, quantum machine learning offers a promising avenue for future research. The study suggests that further exploration of quantum algorithms and their integration with classical models could lead to more efficient and accurate detection methods. The paper also emphasizes the need for more up-to-date datasets and the development of frameworks for quantum machine learning beyond the current Qiskit library. Overall, the study provides valuable insights into the potential of quantum machine learning in the field of cybersecurity.This paper presents a comprehensive study on the detection of malicious URLs using both classical and quantum machine learning (QML) techniques. The research focuses on identifying fraudulent URLs that mimic legitimate ones, which are commonly used in phishing attacks. The study evaluates various machine learning algorithms, including logistic regression, decision trees, support vector machines (SVM), and neural networks, to achieve high accuracy in detecting malicious URLs. The results show that these classical methods can achieve true positive rates greater than 90% when applied to different datasets. The study also explores the application of quantum machine learning for URL detection. Quantum machine learning offers a promising approach by leveraging quantum computing principles, such as superposition and entanglement, to potentially improve the efficiency and accuracy of classification models. The research introduces a Variational Quantum Classifier (VQC), which combines quantum circuits with classical optimization techniques to perform binary classification of URLs. The VQC was tested on a dataset of URLs, and the results showed that it could achieve high accuracy, although it was found that classical models generally outperformed quantum models in this specific task. The study highlights the importance of data preprocessing, including feature selection and encoding, to ensure the effectiveness of both classical and quantum models. The dataset used in the study contains information about URLs, including their length, content, and server type, and was preprocessed to remove irrelevant features and balance the dataset. The results indicate that the dataset is unbalanced, with a higher number of fraudulent URLs than non-fraudulent ones, which affects the performance of the models. The research concludes that while classical machine learning models are currently more effective in detecting malicious URLs, quantum machine learning offers a promising avenue for future research. The study suggests that further exploration of quantum algorithms and their integration with classical models could lead to more efficient and accurate detection methods. The paper also emphasizes the need for more up-to-date datasets and the development of frameworks for quantum machine learning beyond the current Qiskit library. Overall, the study provides valuable insights into the potential of quantum machine learning in the field of cybersecurity.