Skip to content

cja5553/tweets_mining_social_concerns_orphans

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

53 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Title:

Submitted to Children and Youth Services Review

Please Cite as: Lessons from #COVID-19 Tweets: Using social media mining to understand the preventable shortfalls in services to orphans in future crises

Abstract

In the wake of global emergencies like the COVID-19 pandemic, it becomes imperative to refine social support systems that were designed to protect vulnerable children, such as those for orphans, to prevent the recurrence of shortcomings in future crises. However, there is a deficit in literature regarding the pandemic’s repercussions on orphan care, owing to the challenges of conducting traditional, survey-based research during such crises. This study innovatively leverages social media mining to bridge this gap. Through the deployment of text-mining tools such as topic modeling, sentiment analysis, and emotional analysis on COVID-19-related tweets mentioning orphans’, we were able to gather insights into the public’s perception of the social conditions faced by orphans. Our results demonstrate that tweets containing the emotions of anger’ and ‘disgust’ showed a significant increase throughout the progression of the COVID-19 pandemic. Additionally, results from our topic models reveal several concerns regarding orphans during the COVID-19 pandemic: the pressing need for food and care, firsthand trauma experienced by orphans, and a tragic increase in newlyorphaned children who were left homeless as a result of COVID-19. By integrating insights from emotion classification and topic modeling with existing literature, we establish a critical foundation for proposing strategies to strengthen systems designed to protect orphans in future crises.

Research Design & Methods

Data Cleaning --> Text Analysis (Topic Modeling & Sentiment/Emotion classification) --> Regression Analysis (includes visualization)

Data

The "Coronavirus (covid19) Tweets dataset" is publicly available at https://www.kaggle.com/datasets/smid80/coronavirus-covid19-tweets. Due to Twitters terms & conditions, the data cannot be shared via this repository. Please refer to the link for the dataset.

The “John Hopkins University Center for Systems Science and Engineering COVID-19 Data” is publicly available at https://ourworldindata.org/coronavirus. Due to owid's terms & conditions, the data cannot be shared via this repository. Please refer to the link for the dataset.