Beyond classical approaches: redefining the landscape of high-accurate movie recommendation using QNN

Recommender systems are crucial in delivering personalized content and enhancing user satisfaction. This paper investigates the performance of various machine learning algorithms, classical neural networks (CNNs), and quantum neural networks (QNNs) for movie recommendations using the MovieLens-1 M dataset. Traditional approaches like random forest, K-means, and support vector machines (SVM) are evaluated against the newly proposed QNN models. Two distinct QNN architectures are introduced, leveraging the principles of quantum computing, such as superposition and entanglement, to enhance recommendation accuracy. The study demonstrates that the simple QNN architecture significantly outperforms traditional machine learning models and CNNs, reducing prediction errors by 6% in terms of Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The complex QNN model, while accurate, requires more computational resources compared to its simpler counterpart. This research highlights the transformative potential of quantum computing in recommender systems, offering unprecedented accuracy and personalization. Future research will focus on scalability, practical implementation of QNNs, and exploring their application across diverse domains.