In this study, researchers created a natural language processing (NLP) framework to analyse the perspectives on HPV vaccination expressed on Twitter during a 10-year period, from 2008 to 2017. The sentiment study shows how sentiment has changed over the last ten years. According to the findings, there are more unfavourable tweets from 2008 to 2011 and from 2015 to 2016. Entity extraction and analysis aid in the identification of organisations, geographical locations, and events entities linked with negative and positive tweets. The findings reveal that organisation entities such as FDA, CDC, and Merck appear in both negative and positive tweets virtually every year, but geographical location entities cited in both negative and positive tweets vary from year to year. The reason for this is because of distinct events that occurred in those various areas. The goal of AI-based phrase association mining is to discover the major issues represented in both negative and positive tweets, as well as comprehensive tweet content.

They discovered that the most popular negative subjects on Twitter are “injuries,” “deaths,” “scandal,” “safety concerns,” and “adverse/side effects,” whereas the most popular positive topics are “cervical cancers,” “cervical screens,” “prevents,” and “vaccination campaigns.” They think that the findings of this study will assist public health experts in better understanding the nature of social media effect on HPV vaccination attitudes and in developing methods to combat the spread of misinformation.