Abstract:
A huge amount of potentially dangerous COVID-19 misinformation is appearing online. Here we use machine learning to quantify COVID-19 content among online opponents of establishment health guidance, in particular vaccinations (“anti-vax”). We find that the anti-vax community is developing a less focused debate around COVID-19 than its counterpart, the pro-vaccination (“pro-vax”) community. However, the anti-vax community exhibits a broader range of “flavours” of COVID-19 topics, and hence can appeal to a broader cross-section of individuals seeking COVID-19 guidance online, e.g. individuals wary of a mandatory fast-tracked COVID-19 vaccine or those seeking alternative remedies. Hence the anti-vax community looks better positioned to attract fresh support going forward than the pro-vax community. This is concerning since a widespread lack of adoption of a COVID-19 vaccine will mean the world falls short of providing herd immunity, leaving countries open to future COVID-19 resurgences. We provide a mechanistic model that interprets these results and could help in assessing the likely efficacy of intervention strategies. Our approach is scalable and hence tackles the urgent problem facing social media platforms of having to analyze huge volumes of online health misinformation and disinformation.
INDEX TERMS COVID-19, machine learning, topic modelling, mechanistic model, social computing.
Existing System:
Unlike many existing works, this study does not use Twitter data [16][17] since it is known that Twitter is more of a broadcast medium for individual shout-outs, whereas discussions and narratives tend to be nurtured in in-built online community spaces that are a specific feature of platforms like Facebook (e.g., fan page) [18]. Twitter does not have such in-built community spaces. In the present methodology, generalized from Refs. [19] and [20], data is collected from these online communities, specifically Facebook Pages that support either anti-vaccination or pro-vaccination views. This information is publicly available and does not require any individual details, thereby avoiding any privacy concerns – just as understanding the content of a conversation among a crowd of people in an open, real-world public space does not require knowledge of any personal details about the individuals within that crowd. Details of our approach are given in Sec. II and the Appendix. A third difference between this study and previous ones is that the machine learning findings here are interpreted in terms of a mechanistic model (Sec. IV) that captures the general trend for the coherence in the online conversations over time. Though much work still needs to be done, this study therefore provides a first step toward a fully automated but interpretable understanding of the growing public health debate concerning vaccines and COVID-19.
Proposed System:
Machine learning automation can, in principle, help address the significant issues facing social media platforms by mechanically picking out material that requires attention from the huge haystack of online content. While this could help to better curtail online misinformation, one might rightly ask about its accuracy and reliability as compared to human analysts. This has been recently addressed in Ref. [24]. We use the same coherence metric (CV) as these authors. They addressed the problem that topic models had previously given no guarantee on the interpretability of their output. Specifically, they produced several benchmark datasets with human judgements of the interpretability of topics and they found results that outperformed existing measures with respect to correlation to human ratings. They achieved this by evaluating 237,912 coherence measures on 6 different benchmarks for topic coherence, making this the biggest study of topic coherences at that time. Separately, we have done our own comparison for the general area of online hate and have found comparable consistency.
SYSTEM REQUIREMENTS
SOFTWARE REQUIREMENTS:
• Programming Language : Python
• Font End Technologies : TKInter/Web(HTML,CSS,JS)
• IDE : Jupyter/Spyder/VS Code
• Operating System : Windows 08/10
HARDWARE REQUIREMENTS:
Processor : Core I3
RAM Capacity : 2 GB
Hard Disk : 250 GB
Monitor : 15″ Color
Mouse : 2 or 3 Button Mouse
Key Board : Windows 08/10
