JMIR Publications

ABSTRACT

Background:

Given the limitations of medical diagnosis of early emotional change signs during the COVID-19 quarantine period, artificial intelligence models provide effective mechanisms in uncovering early signs, symptoms and escalating trend.

Objective:

The main purpose of this project is to demonstrate the effectiveness of Artificial Intelligence, and in particular Natural Language Processing and Machine Learning in detecting and analyzing emotions from tweets talking about COVID-19 social confinement.

Methods:

We developed a systematic framework that can be directly applied to COVID-19 related mood discovery, using eight types of emotional reaction and designing a deep learning model to uncover emotions based on the first wave of the pandemic public health restriction of mandatory social segregation. We argue that the framework can discover semantic trends of COVID-19 tweets during the first wave of the pandemic to predict new concerns that may be associated with furthering into the new waves of COVID-19 quarantine orders and other related public health regulations.

Results:

Our findings revealed Stay-At-Home restrictions result in people expressing on twitter both negative and positive based on emotional and semantics aspects. Moreover, the statistical results of the emotion classification is show that our framework based on CNN deep learning has predicted the emotion levels or target labels with more F1-socore than the LSTM model, which are 0.95% and 0.93%, respectively. However, these results have potential to impact public health policy decisions through monitoring trends of emotional feelings of those who are quarantined.

Conclusions:

The research shows that the framework is effective in capturing the emotions and semantics trends in social media messages during the pandemic. Moreover, the framework can be applied to uncover reactions to similar public health policies that affect people’s well-being.