Aspect based sentiment analysis using multi-criteria decision-making and deep learning under COVID-19 pandemic in...

Area of Science:

• Social Sciences
• Computer Science
• Public Health

Background:

• The COVID-19 pandemic significantly impacted global health and economies, leading to widespread lockdowns.
• Social media platforms like Twitter became crucial for expressing public emotions and opinions during these restrictions.
• Understanding public sentiment is vital for evaluating the effectiveness of public health interventions.

Purpose of the Study:

• To analyze Indian public sentiment towards the Government of India's COVID-19 lockdown measures.
• To investigate public feelings during both lockdown and stepwise unlock phases using Twitter data.
• To apply Aspect-Based Sentiment Analysis for detailed opinion mining.

Main Methods:

• Collected and annotated India-specific COVID-19 tweets.
• Utilized Aspect-Based Sentiment Analysis to determine public opinion polarity.
• Developed and applied a deep learning model for classifying Twitter data sentiment.
• Compared the proposed model's performance against traditional methods like LSTM, Bi-LSTM, and GRU.

Main Results:

• The deep learning model achieved high accuracy: 82.35% for the lockdown dataset and 83.33% for the unlock dataset.
• The proposed method demonstrated superior performance compared to several contemporary approaches.
• Public sentiment analysis provided insights into various aspects of the lockdown and unlock phases.

Conclusions:

• The study successfully highlighted public sentiment regarding the Indian government's COVID-19 lockdown and unlock strategies.
• The deep learning approach proved effective in analyzing large-scale social media data for public opinion.
• Findings offer valuable insights for policymakers on public perception of pandemic control measures.