Social media polarization is an uncertain risk to any business, and mitigating this becomes challenging for organizations. In this work, we investigate the formation of opinion polarization on social media by using Natural Language Processing during COVID 19 pandemic. The wheel of emotions theory is used to quantify the opinions with the help of eight basic emotions. Topic modeling of those opinions indicates the issues discussed by the crowd over social media. The opinions are clustered on the basis of the emotion-based similarity measure. The existence of topics in exactly opposite emotion clusters indicates opinion polarization over a particular issue. The strength of polarization has been identified using the emotional intensity of the topic. This study proposes a mechanism to identify opinion polarization in social media and gives a cue to business professionals indicating a significant potential future risk that might need necessary mitigation measures
Dey, Debasmita and Kumar, Pradeep, "Investigation of Social Media Polarization using Natural Language Processing: A Case of COVID 19 Pandemic" (2021). PACIS 2021 Proceedings. 90.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.