Exploring Internet Meme Activity during COVID-19 Lockdown Using Artificial Intelligence Techniques

Figures & data

Figure 1. Overall methodology of the study.

Figure 2. Popular memes considered for the study.

Figure 3. (a-d) Number of Captions over time for Internet Memes (individual graphs)

Figure 3. Continued.

Figure 4. Line plots depicting number of captions over time for internet memes.

Figure 5. Line plots depicting number of captions over time for internet memes.

Figure 6. Overall internet meme trend (September 2017- August 2020).

Figure 7. Heat map with 8 clusters.

Figure 8. Heat map with 15 clusters.

Figure 9. RF forecast.

Figure 10. IL forecast.

Figure 11. MLP forecast.

Figure 12. Classification accuracy.

Figure 13. Kappa statistic.

Figure 14. MAE and RMSE plots.

Table 1. Comparative analysis of our work with existing works

Author and Year	Analysis/ Study Performed	Methodology/ Parameters	Results
(Elmer, Mepham, and Stadtfeld 2020)	Social network activity and changes in mental health	Web-Based Questionnaire, Social Networks (September 2018- April 2020)	Increased Social Network Activity, Worries. Isolation, Lack of emotional support, stress.
(Kochhar et al. 2020)	Survey on Mental Health	Survey and Web-Based Questionnaire	Lockdown is a predominant stressor for mental health.
(Thakur and Jain 2020)	Analyzing global suicides	Survey and Media Reports	Increased number of global suicides
(Singh et al. 2020)	Analyzing pollution levels	Estimating particulate matter and pollutants (March 25, 2020 – May 03, 2020) across India	Improved air quality due to changes in emissions and meteorology
(Colley, Bushnik, and Langlois 2020)	Study on exercise schedule and screen time	Web-Based Survey	Increase in indoor exercise. Increased TV time, internet use, and video game time.
(Ni et al. 2020)	Analyzing mental health and social media behavior	Online survey via WeChat	Results indicated anxiety and depression in subjects. Increased use of social media (~2 hours)
(Feldmann et al. 2021a)	Analyzing Internet Traffic during COVID-19 pandemic	Internet traffic data collected from service providers, exchange points and academic networks	15–20% increase in internet activity in first few weeks (in some cases 200%)
(Candela, Luconi, and Vecchio 2020)	Analyzing Internet latency during COVID-19 pandemic	Studying packet delay on internet Data collected from RIPE Atlas	Delay is 3–4 times more during the pandemic. Packet loss is 2–3 times more.
(Feldmann et al, 2021a)	Impact of pandemic on internet traffic	Data collected from service providers and operators. Traffic volume and usage patterns analyzed.	Traffic volume increased by 40%, traffic usage increased upto 20%
(Bhatti et al. 2020)	Analyzing e-Commerce during lockdown	Number of people shopping on retail websites	4059 million people (Amazon), followed by 1227 million people (Ebay) and 648 million people (Samsung)
Our Proposed work	Analyzing Internet meme activity over social media during COVID-19 lockdown	Artificial Intelligence techniques like RF, MLP, IL algorithms	Evaluation techniques like RMSE, MAE, Accuracy and Kappa statistics. Accuracy is nearly 99%.

Elmer, T., K. Mepham, and C. Stadtfeld. 2020. Students under lockdown: Comparisons of students’ social networks and mental health before and during the COVID-19 crisis in Switzerland. PloS One 15 (7):e0236337. doi:10.1371/journal.pone.0236337.

 PubMed Web of Science ®Google Scholar

Kochhar, A. S., R. Bhasin, G. K. Kochhar, H. Dadlani, V. V. Mehta, R. Kaur, and C. K. Bhasin. 2020. Lockdown of 1.3 billion people in India during Covid-19 pandemic: A survey of its impact on mental health. Asian Journal of Psychiatry 54:102213. doi:10.1016/j.ajp.2020.102213.

 PubMed Web of Science ®Google Scholar

Thakur, V., and A. Jain. 2020. COVID 2019-suicides: A global psychological pandemic. Brain, behavior, and immunity. https://www.sciencedirect.com/science/article/pii/S0889159120306437?via%3Dihub

 Google Scholar

Singh, V., S. Singh, A. Biswal, A. P. Kesarkar, S. Mor, and K. Ravindra. 2020. Diurnal and temporal changes in air pollution during COVID-19 strict lockdown over different regions of India, 115368. Environmental Pollution.

 Google Scholar

Colley, R. C., T. Bushnik, and K. Langlois. 2020. Exercise and screen time during the COVID-19 pandemic. Health Rep 31 (6):3–11.

 Web of Science ®Google Scholar

Ni, M. Y., L. Yang, C. M. Leung, N. Li, X. I. Yao, Y. Wang, … Q. Liao. 2020. Mental health, risk factors, and social media use during the COVID-19 epidemic and cordon sanitaire among the community and health professionals in Wuhan, China: Cross-sectional survey. JMIR Mental Health 7 (5):e19009. doi:10.2196/19009.

 PubMed Web of Science ®Google Scholar

Feldmann, A., O. Gasser, F. Lichtblau, E. Pujol, I. Poese, C. Dietzel, … G. Smaragdakis. 2021a. March. Implications of the COVID-19 Pandemic on the Internet Traffic. In Broadband Coverage in Germany; 15th ITG-Symposium. VDE, 1–5, Germany.

 Google Scholar

Candela, M., V. Luconi, and A. Vecchio. 2020. Impact of the COVID-19 pandemic on the Internet latency: A large-scale study. Computer Networks 182:107495. doi:10.1016/j.comnet.2020.107495.

 PubMed Web of Science ®Google Scholar

Bhatti, A., H. Akram, H. M. Basit, A. U. Khan, S. M. Raza, and M. B. Naqvi. 2020. E-commerce trends during COVID-19 Pandemic. International Journal of Future Generation Communication and Networking 13 (2):1449–52.

 Google Scholar