A CNN-LSTM-Based Hybrid Deep Learning Approach for Sentiment Analysis on Monkeypox Tweets

The research on sentiment analysis has shown a great deal of utility in the field of public health, specifically in the investigation of infectious illnesses. As the world begins to recuperate from the devastating effects of the COVID-19 pandemic, there is a growing concern that a different pandemic, known as Monkeypox, may strike the world once more. The contagious illness known as Monkeypox has been documented in over 73 countries worldwide. This unexpected epidemic has become a significant cause of anxiety for many people and health authorities. Various social media platforms have presented various perspectives regarding the monkeypox epidemic. Our goal is to research how the public feels about the recent Monkeypox epidemic to assist policymakers in developing a deeper comprehension of how the public views the illness. This research uses a CNN-LSTM-based hybrid architecture to ascertain people's feelings regarding Monkeypox disease. A series of experiments were conducted on an open-access dataset of tweets related to the Monkeypox. The tweets undergo various pre-processing, global vectorization, and one-hot encoding techniques. According to the findings of our experiments, the hybrid model provided better accuracy, which was approximately 91%. In addition, the findings are validated by contrasting them with more conventional machine learning techniques. The outcomes of this investigation contribute to a general population that has a greater awareness of the Monkeypox infection.