Emotional Analysis using Twitter Data during Pandemic Situation: COVID-19

Abstract

During the crisis situation caused due to COVID-19 disease, managing mental health and psychological well-being is as important as physical health of people. As web based life is broadly utilized by individuals to communicate their feeling and supposition, our framework utilizes Twitter information posted by individuals during this emergency circumstance to dissect the feelings of individuals. For processing the cleaned data NRC Word-Emotion Association Lexicon (have aka EmoLex) is used. NRC Word-Emotion Association Lexicon is a list of English with real-valued scores of intensity for eight basic emotion words ns (anger, anticipation, disgust, fear, joy, sadness, surprise, and trust). The text content of tweeter dataset created by fetching tweets across the world have classified into basic emotions like anger, anticipation, disgust, fear, joy, sadness, surprise and trust. This analysis can be used by authorities to understand the mental health of the people and can take necessary measures to decide on policies to fight against coronavirus which is affecting the social well-being as well as economy of the whole world.

Covid-19 Discourse on Twitter: How the Topics, Sentiments, Subjectivity, and Figurative Frames Changed Over Time

“…However, to the best of our knowledge, they appear to use a significantly different methodology. Those works include Sentiment Analysis with Deep Learning Classifiers (Chakraborty et al, 2020;Li et al, 2020), a time-span of much fewer days (Abd-Alrazaq et al, 2020;Xue et al, 2020), and analyses of specific emotions without topic models (Lwin et al, 2020;Mathur et al, 2020). Because topic modeling is an exploratory, bottom-up, data-driven technique of data mining, we believe that a broader and more explorative approach, that takes into account multiple topic modeling solutions and a longer time span, may provide better insights on the themes discussed by Twitter users over time.…”

“…In the case of political messaging during electoral campaigns, positive feedback might correlate with voters' support for a specific candidate. In some cases, as pointed out by recent studies, social media analyses during crisis situations may be used to investigate real-time public opinion and thus help authorities to gain insight for quickly deciding for the best assistance policies to be taken (Mathur et al, 2020).…”

Perbandingan Performansi Model pada Algoritma K-NN terhadap Klasifikasi Berita Fakta Hoaks Tentang Covid-19

“…Setelah World Health Organization (WHO) menyatakan Corona Virus Disease of 2019 (Covid-19) sebagai wabah pandemi pada awal tahun 2020, penggunaan dan penerapan aplikasi daring bertambah digunakan dalam berbagai bidang untuk mencegah penyebaran Covid-19 karena dapat menjaga jarak atau mengurangi kerumunan. Selama pandemi Covid-19 berlangsung terdapat penyebaran informasi tidak tepat (Mathur, Kubde, & Vaidya, 2020) . Berdasarkan informasi palsu yang disampaikan, terdapat pengaruh dari penyampaian berita hoaks tersebut mengenai virus Covid-19 yang terdapat pada internet terhadap pembentukan opini masyarakat (Roy & Junaidi, 2020) dengan persentase sebesar 58,7% yang dapat mempengaruhi pengguna internet dan 41,3% lainnya dipengaruhi oleh faktor lain.…”

Social and institutional presence of the Heads of Government of the Americas on Social Media

“…Castells (2017) gives examples of the Arab Spring demonstrations in 2010 and the Occupy Wall Street movement in 2011. Mathur, Kubde, and Vaidya (2020) used Twitter data to analyze across the world the mental health of people during the COVID-19 pandemic situation through emotion analysis and classified it into basic emotions. The idea is via their analysis, authorities can better understand the mental health of the people and update the content accordingly.…”