| COVID-19 detection [29]–[31]
|
Combined COVID-19 Dataset [32] and Noisy COVID-19 X-ray Dataset [28]
|
Machine Learning and Deep Learning models |
LWL-SOM performed better than the current machine learning models on this data |
| Bibiliometric analysis of COVID-19 [33]–[36]
|
COVID-19 bibliometric data |
Packages like VOSviewer, Bibliometrix and R software |
Identified some research themes but unable to address other themes like the long-term impact |
| Textual analysis of COVID-19 tweets [37]
|
COVID-19 tweet data |
Topic Modeling, UMAP, and DiGraphs |
Identified the topics, key terms, and features of COVID-19 tweets |
| Comparative analysis of COVID-19 research with respect to other coronavirus research [33], [38], [39]
|
SARS, MERS and COVID-19 publications dataset |
Packages like VOSviewer, biblioshiny and R software |
Identified the influential aspects of different areas in coronavirus research |
| Topic Modeling: COVID-19 research with respect to other coronavirus research [40], [41]
|
SARS, MERS, COVID-19 and other CoV publications (CORD-19) dataset |
Latent Dirichlet Allocation |
Identified inter-relationships between various coronavirus research to address knowledge gaps |
