Feature-Based Sentimental Analysis on Public Attention towards COVID-19 Using CUDA-SADBM Classification Model.

Area of Science:

• Computational Social Science
• Artificial Intelligence
• Public Health Informatics

Background:

• The COVID-19 pandemic has caused widespread mental and economic distress globally.
• Accurate interpretation of public sentiment is crucial for understanding pandemic impact and public health responses.
• Existing methods may not fully capture the nuances of public opinion expressed on social media.

Purpose of the Study:

• To develop a robust model for analyzing public sentiment regarding the COVID-19 pandemic.
• To leverage social media data for more accurate pandemic statistics and public opinion insights.
• To introduce and evaluate a novel Sentimental DataBase Miner algorithm (SADBM).

Main Methods:

• Sentiment analysis using a unique classifier, the Sentimental DataBase Miner algorithm (SADBM).
• Data collection from various online social media platforms including Twitter, Facebook, and LinkedIn.
• Comparative analysis against basic classifiers like logistic regression and decision tree.
• Performance evaluation on both CPU and GPU, including calculation of acceleration ratio.

Main Results:

• The proposed SADBM model demonstrated superior accuracy in sentiment categorization compared to logistic regression and decision tree.
• The SADBM algorithm achieved a high accuracy of 96% when executed on a Graphics Processing Unit (GPU).
• Parallel processing on GPU significantly accelerated the model's execution.

Conclusions:

• The SADBM algorithm is an effective tool for sentiment analysis of public opinion during health crises like the COVID-19 pandemic.
• GPU acceleration enhances the efficiency of sentiment analysis models for large-scale social media data.
• Accurate sentiment analysis can provide valuable insights for public health monitoring and response strategies.