COVIDSenti: A Large-Scale Benchmark Twitter Data Set for COVID-19 Sentiment Analysis

Abstract

Social media (and the world at large) have been awash with news of the COVID-19 pandemic. With the passage of time, news and awareness about COVID-19 spread like the pandemic itself, with an explosion of messages, updates, videos, and posts. Mass hysteria manifest as another concern in addition to the health risk that COVID-19 presented. Predictably, public panic soon followed, mostly due to misconceptions, a lack of information, or sometimes outright misinformation about COVID-19 and its impacts. It is thus timely and important to conduct an ex post facto assessment of the early information flows during the pandemic on social media, as well as a case study of evolving public opinion on social media which is of general interest. This study aims to inform policy that can be applied to social media platforms; for example, determining what degree of moderation is necessary to curtail misinformation on social media. This study also analyzes views concerning COVID-19 by focusing on people who interact and share social media on Twitter. As a platform for our experiments, we present a new large-scale sentiment data set COVIDSENTI, which consists of 90 000 COVID-19-related tweets collected in the early stages of the pandemic, from February to March 2020. The tweets have been labeled into positive, negative, and neutral sentiment classes. We analyzed the collected tweets for sentiment classification using different sets of features and classifiers. Negative opinion played an important role in conditioning public sentiment, for instance, we observed that people favored lockdown earlier in the pandemic; however, as expected, sentiment shifted by mid-March. Our study supports the view that there is a need to develop a proactive and agile public health presence to combat the spread of negative sentiment on social media following a pandemic.

Deep Learning‐Based Methods for Sentiment Analysis on Nepali COVID‐19‐Related Tweets

“…” On the other hand, only 7% are related to “relieved mood. ” Similarly, Naseem et al [ 5 ] compared the traditional machine learning methods such as support vector machine (SVM), Naive Bayes (NB), decision tree (DT), random forest (RF), and deep learning methods such as convolution neural network (CNN), and bidirectional long short-term memory (BiLSTM), in combination with various embedding vectors such as fastText [ 19 ], Glove [ 23 ], and Word2Vec [ 24 ] on their COVID-19 tweets' datasets into three classes (negative, positive, and neutral). Their results depict that the deep learning (DL)-based methods outperform the traditional ML methods.…”

“…There are three main limitations in the aforementioned works. First, most of the existing works [ 2 , 4 , 5 , 12 ] on COVID-19-related tweets are performed in high-resource languages such as English and Arabic. The approach used by high-resource language might be inapplicable to low-resource languages such as Nepali, which is based on Devanagari script and has 36 consonants (33 are distinct consonants and 3 are combined consonants), 13 vowels, and 10 numerals ( Figure 1 ) [ 1 , 15 , 16 ].…”

COVID-19 Vaccination Awareness and Aftermath: Public Sentiment Analysis on Twitter Data and Vaccinated Population Prediction in the USA

“…Machine learning (ML) is the most recent method in data science that has paved the way for technological accomplishments and tools that would have been unimaginable a couple of years ago. Image recognition, sentiment analysis [1][2][3][4], product recommendations, spam/fraud detection [5], social media features, etc. are some of the real-world machine learning applications that are sweeping the world.…”

Analysis of lockdown perception in the United States during the COVID-19 pandemic

“…: V,-vol extract emotions conveyed by selected texts [34][35][36][37][38][39], in terms of a discrete classification or a continuous score. Twitter data on COVID-19 pandemic has been used to study reactions to the outbreak in different countries [40][41][42], benchmark and validate new models for natural language processing [43][44][45], perform sentiment analysis about the pandemic [46][47][48], and conduct analyses about a specific event [49].…”