An optimistic firefly algorithm-based deep learning approach for sentiment analysis of COVID-19 tweets

Area of Science:

• Computational linguistics
• Artificial intelligence
• Public health informatics

Background:

• The COVID-19 pandemic significantly impacted global health and well-being.
• Social media platforms, particularly Twitter, became crucial for public discourse and sentiment expression during the pandemic.
• Monitoring public sentiment is vital for understanding the psychological impact and managing the spread of infectious diseases.

Purpose of the Study:

• To develop and evaluate a deep learning model for analyzing public sentiment on COVID-19 from Twitter data.
• To enhance the sentiment analysis model's performance using the firefly algorithm.
• To compare the proposed model's effectiveness against existing machine learning and ensemble methods.

Main Methods:

• Utilized a Long Short-Term Memory (LSTM) deep learning network for sentiment analysis of COVID-19 related tweets.
• Integrated the firefly algorithm to optimize the LSTM model's parameters and improve performance.
• Evaluated the model using standard performance metrics including accuracy, precision, recall, AUC-ROC, and F1-score.

Main Results:

• The proposed LSTM + Firefly model achieved a high accuracy of 99.59% in classifying tweet sentiments.
• Demonstrated superior performance compared to other state-of-the-art ensemble and machine learning models.
• The firefly algorithm significantly enhanced the sentiment analysis capabilities of the LSTM model.

Conclusions:

• The LSTM + Firefly approach is a highly effective method for analyzing public sentiment regarding COVID-19 on social media.
• This technique offers a valuable tool for public health officials to monitor and understand public perception during health crises.
• Accurate sentiment analysis can aid in developing targeted interventions and communication strategies.