Proposed Model for QCNN-Based Sentimental Short Sentences Classification

As social networking continues to expand, web users have been sharing their thoughts and viewpoints daily, utilizing various mediums such as texts, images, videos, and speech. However, despite this active participation, text classification remains a crucial challenge due to the sheer volume of texts received from diverse sources and individuals with different mindsets. The shared opinions often prove to be incomplete, inconsistent, and noisy, further complicated by variations in languages. To address these challenges, NLP (Natural Language Processing) and Quantum Machine Learning (QML) methods have become widely employed. This study focuses on exploring the potential of current quantum computers in enhancing the performance of natural language processing tasks. Specifically, we propose a new approach called the Quantum Convolutional Neural Network (QCNN) for sentiment analysis. Our proposed model is the first model based on QCNN at text classification field; it leverages QCNN to extract more effective features from short sentences; Thereby, improving sentiment analysis accuracy and efficiency.