Nowcasting unemployment rate during the COVID-19 pandemic using Twitter data: The case of South Africa

The global economy has been hard hit by the COVID-19 pandemic. Many countries are experiencing a severe and destructive recession. A significant number of firms and businesses have gone bankrupt or been scaled down, and many individuals have lost their jobs. The main goal of this study is to support policy- and decision-makers with additional and real-time information about the labor market flow using Twitter data. We leverage the data to trace and nowcast the unemployment rate of South Africa during the COVID-19 pandemic. First, we create a dataset of unemployment-related tweets using certain keywords. Principal Component Regression (PCR) is then applied to nowcast the unemployment rate using the gathered tweets and their sentiment scores. Numerical results indicate that the volume of the tweets has a positive correlation, and the sentiments of the tweets have a negative correlation with the unemployment rate during and before the COVID-19 pandemic. Moreover, the now-casted unemployment rate using PCR has an outstanding evaluation result with a low Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), Symmetric MAPE (SMAPE) of 0.921, 0.018, 0.018, respectively and a high R2-score of 0.929.

[1]  Vinit Kumar Gunjan,et al.  IoT enabled HELMET to safeguard the health of mine workers , 2022, Comput. Commun..

[2]  A. R. Javed,et al.  A survey on COVID-19 impact in the healthcare domain: worldwide market implementation, applications, security and privacy issues, challenges and future prospects , 2022, Complex & Intelligent Systems.

[3]  A. R. Javed,et al.  Exploratory data analysis, classification, comparative analysis, case severity detection, and internet of things in COVID-19 telemonitoring for smart hospitals , 2022, J. Exp. Theor. Artif. Intell..

[4]  Siddhaling Urolagin,et al.  Deep Learning-based Sentiment Analysis and Topic Modeling on Tourism During Covid-19 Pandemic , 2021, Frontiers in Computer Science.

[5]  Antonio Javier Gallego,et al.  A multimodal approach for regional GDP prediction using social media activity and historical information , 2021, Appl. Soft Comput..

[6]  Nasir Imran,et al.  Extreme Value Analysis and Risk Assessment: A Case of Pakistan Stock Market , 2021, ECONOMIC COMPUTATION AND ECONOMIC CYBERNETICS STUDIES AND RESEARCH.

[7]  Rachel E White,et al.  Child Adjustment During COVID-19: The Role of Economic Hardship, Caregiver Stress, and Pandemic Play , 2021, Frontiers in Psychology.

[8]  Z. Satti,et al.  Examining investors' sentiments, behavioral biases and investment decisions during COVID-19 in the emerging stock market: a case of Pakistan stock market , 2021, Journal of Economic and Administrative Sciences.

[9]  Matthijs J. Warrens,et al.  The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation , 2021, PeerJ Comput. Sci..

[10]  Leticia Elizabeth Romero-García,et al.  Social network analysis of spreading and exchanging information on Twitter: the case of an agricultural research and education centre in Mexico , 2021, The Journal of Agricultural Education and Extension.

[11]  Clement Ola Adekoya,et al.  Social media and the spread of COVID-19 infodemic , 2021 .

[12]  George Alexander,et al.  Automating Large-scale Health Care Service Feedback Analysis: Sentiment Analysis and Topic Modeling Study , 2021, JMIR medical informatics.

[13]  H. Beazley,et al.  Children during the COVID-19 pandemic: children and young people’s vulnerability and wellbeing in Indonesia , 2021, Children's Geographies.

[14]  Emmanouil I. Marakakis,et al.  OpinionMine: A Bayesian-based framework for opinion mining using Twitter Data , 2021 .

[15]  Samah J. Fodeh,et al.  Twitter-based analysis reveals differential COVID-19 concerns across areas with socioeconomic disparities , 2021, Computers in Biology and Medicine.

[16]  B. Rimé,et al.  Don’t put all social network sites in one basket: Facebook, Instagram, Twitter, TikTok, and their relations with well-being during the COVID-19 pandemic , 2021, PloS one.

[17]  E. Anagnostou,et al.  Mostly worse, occasionally better: impact of COVID-19 pandemic on the mental health of Canadian children and adolescents , 2021, European Child & Adolescent Psychiatry.

[18]  Md. Mokhlesur Rahman,et al.  Socioeconomic factors analysis for COVID-19 US reopening sentiment with Twitter and census data , 2021, Heliyon.

[19]  Daniele Toninelli,et al.  Comparing Methods to Collect and Geolocate Tweets in Great Britain , 2021 .

[20]  Rodrigo Sandoval-Almazán,et al.  Does Twitter Affect Stock Market Decisions? Financial Sentiment Analysis During Pandemics: A Comparative Study of the H1N1 and the COVID-19 Periods , 2021, Cogn. Comput..

[21]  Benjamin K. P. Woo,et al.  Twitter as a Mental Health Support System for Students and Professionals in the Medical Field , 2021, JMIR medical education.

[22]  Joanna Michalak.  Does pre-processing affect the correlation indicator between Twitter message volume and stock market trading volume? , 2020 .

[23]  Ram Krishn Mishra,et al.  Data Analysis of Novel Coronavirus Based on Multiple Factors , 2020, Information Technology Trends.

[24]  Xiao Huang,et al.  Twitter reveals human mobility dynamics during the COVID-19 pandemic , 2020, PloS one.

[25]  Philip S. Yu,et al.  Understanding Pre-trained BERT for Aspect-based Sentiment Analysis , 2020, COLING.

[26]  S. Boon-itt,et al.  Public Perception of the COVID-19 Pandemic on Twitter: Sentiment Analysis and Topic Modeling Study , 2020, JMIR public health and surveillance.

[27]  Leonardo Neves,et al.  TweetEval: Unified Benchmark and Comparative Evaluation for Tweet Classification , 2020, FINDINGS.

[28]  Yashar Salamzadeh,et al.  Entrepreneurial Orientation and Small and Medium-sized Enterprises’ Performance; Does ‘Access to Finance’ Moderate the Relation in Emerging Economies? , 2020 .

[29]  P. Butterworth,et al.  Unemployment, Employability and COVID19: How the Global Socioeconomic Shock Challenged Negative Perceptions Toward the Less Fortunate in the Australian Context , 2020, Frontiers in Psychology.

[30]  Mark Dredze,et al.  The Twitter Social Mobility Index: Measuring Social Distancing Practices With Geolocated Tweets , 2020, Journal of medical Internet research.

[31]  P. Patnaik,et al.  COVID ‐19 pandemic! It's impact on people, economy, and environment , 2020 .

[32]  Rodrigo Sandoval-Almazán,et al.  Does Twitter affect Stock Market Decisions?Financial Sentiment Analysis in Pandemic Seasons: A Comparative Study of H1N1 and COVID-19 , 2020 .

[33]  Alese E. Halvorson,et al.  Well-being of Parents and Children During the COVID-19 Pandemic: A National Survey , 2020, Pediatrics.

[34]  Xiangliang Zhang,et al.  SenWave: Monitoring the Global Sentiments under the COVID-19 Pandemic , 2020, ArXiv.

[35]  Steven J. Davis,et al.  Economic uncertainty before and during the COVID-19 pandemic , 2020, Journal of Public Economics.

[36]  Chen Chen,et al.  Twitter Discussions and Emotions About the COVID-19 Pandemic: Machine Learning Approach , 2020, Journal of medical Internet research.

[37]  Charl Jooste,et al.  Nowcasting Economic Activity in Times of COVID-19: An Approximation from the Google Community Mobility Report , 2020 .

[38]  C. Iglesias,et al.  Predicting Reputation in the Sharing Economy with Twitter Social Data , 2020, Applied Sciences.

[39]  T. Bosch,et al.  Facebook and politics in Africa: Zimbabwe and Kenya , 2020 .

[40]  D. Cucinotta,et al.  WHO Declares COVID-19 a Pandemic , 2020, Acta bio-medica : Atenei Parmensis.

[41]  Namita Srivastava,et al.  The Machine‐Learning Approach , 2020, Machine Learning for iOS Developers.

[42]  Min Song,et al.  Developing a supervised learning-based social media business sentiment index , 2019, The Journal of Supercomputing.

[43]  Pum-Mo Ryu,et al.  Predicting the Unemployment Rate Using Social Media Analysis , 2018, J. Inf. Process. Syst..

[44]  C. Skinner.  Issues and Challenges in Census Taking , 2018 .

[45]  Atsushi Nara,et al.  Understanding the spatio-temporal characteristics of Twitter data with geotagged and non-geotagged content: two case studies with the topic of flu and Ted (movie) , 2017, Ann. GIS.

[46]  Matthias Brenzinger.  Onze llengües oficials i més: legislació i polítiques lingüístiques a Sud-àfrica , 2017 .

[47]  Gábor Vattay,et al.  Prediction of employment and unemployment rates from Twitter daily rhythms in the US , 2017, EPJ Data Science.

[48]  Vlantana Tzinovits.  Using social media to measure labour market flows in Greece , 2016 .

[49]  Scott Counts,et al.  The psychology of job loss: using social media data to characterize and predict unemployment , 2016, WebSci.

[50]  C. R. Nirmala,et al.  Twitter data analysis for unemployment crisis , 2015, 2015 International Conference on Applied and Theoretical Computing and Communication Technology (iCATccT).

[51]  Seokho Lee,et al.  Principal Component Regression by Principal Component Selection , 2015 .

[52]  Douglas Grbic,et al.  Measuring race and ethnicity in the censuses of Australia, Canada, and the United States: Parallels and paradoxes , 2015 .

[53]  Walaa Medhat,et al.  Sentiment analysis algorithms and applications: A survey , 2014 .

[54]  Wenwen Li,et al.  Using geolocated Twitter data to monitor the prevalence of healthy and unhealthy food references across the US , 2014 .

[55]  Michael J. Cafarella,et al.  Using Social Media to Measure Labor Market Flows , 2014 .

[56]  Rajiv T. Maheswaran,et al.  Following Human Mobility Using Tweets , 2012, ADMI.

[57]  Balachander Krishnamurthy,et al.  A few chirps about twitter , 2008, WOSN '08.

[58]  C. Sorrentino.  International Unemployment Rates: How Comparable Are They? , 2000 .

[59]  H. Baxter Williams,et al.  A Survey , 1992 .

[60]  Yogesh K. Dwivedi,et al.  Sentiment analysis and classification of Indian farmers' protest using twitter data , 2021, Int. J. Inf. Manag. Data Insights.

[61]  Ming-Wei Chang,et al.  BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding , 2019, NAACL.

[62]  Masuhiro Kogoma.  総論;総論;Introduction , 2006 .

[63]  Thomas G. Whiston,et al.  A National Survey , 1992 .

[64]  K. Shadan,et al.  Available online: , 2012 .