COVID 19 - The Vaccine Challenge. A twitter text/sentiment analysis - Part 2

In the last few days, there has been a scramble to come up with a vaccine for the deadly Corona Virus. The pandemic to date has seen over 5million cases resulting in 365,015 deaths globally. as reported by John Hopkins.

While a lot is going into the development of a vaccine with top pharmaceutical companies like AstraZeneca leading efforts, however, there seems to be a lot of uncertainty on whether a COVID 19 vaccine would be widely accepted when one eventually becomes available.

Even worrisome are the conspiracy theories that are coming up about vaccines.

A recent AP-Norc poll showed that only half of Americans agree that they will take a COVID 19 Vaccine.

To investigate this further, using R programming language, I performed text and sentiment analysis on twitter data to find out the general perception towards vaccines. My analysis incorporated topic analysis, association analysis to find out popular words that are occurring alongside the word vaccine, and the general sentiment towards vaccines.

The word cloud above shows some of the popular words that came up during my vaccine search.

The following words above appeared frequently 70 or more times when the word "vaccine" was looked up. A few words seemed out of place like "demons" so I applied association analysis to find out what words occurred alongside and why the word "demons" seemed to come up.

Unsurprisingly, I saw words like "Bill gates" and "Melinda" which alluded to some of the bogus conspiracy theories being spread about vaccines. I went on to further leverage topic analysis to get a general perception of conversation topics on vaccines.

Even though I anticipated scores of conversations incorporating vaccine based conspiracy theories, It was great to see that, of the top 12 conversations, only one seemed to center on vaccine conspiracy theories. I conducted an N-gram analysis to further validate this.

To test the general sentiment of people to vaccines and how we should handle the statistics depicted in the AP-Norc Poll, I conducted a sentiment analysis.

Generally from the sentiment analysis, we can largely infer that there is still a high level of trust for vaccines. Also, we see anticipation. We are all eager to return to the world as it was before the pandemic and hopefully, a vaccine can help us achieve this.

Disclaimer: The information depicted may be speculative as it is based on tweets extracted from twitter data. Sentiment analysis is not free of flaws with one of the major shortfalls being its ambiguity - in one situation, a word could be considered positive and at the same time be considered negative in another situation.