Yearly changes in the influenza vaccine has causes mistrust in this important preventative measure. Social media platform, Twitter, has recently been utilised to explore the publics’ opinions and feelings towards different topics through sentiment analysis.
What it does
Who the Flu? is a web application that collects tweets that mention the influenza vaccine, analyses this data using sentiment analysis, and provides a visualisation on the changes in sentiment over time. The user enters the longitude and latitude of the area they would like to explore and the program analyses the Tweets within this area. More negative values indicate negative sentiment whereas more positive values indicate positive sentiment. Ultimately, this information can be used to determine the changes in opinion and will aid in the implementation of public health measures accordingly.
How we built it
The app was built using R and Shiny. The packages used in R allowed us to use the Twitter API as well as perform the sentiment analysis.
Challenges we ran into
This project started off by using the natural language understanding (NLU) function of IBM’s Watson. However, we ran into many issues trying to upload the Twitter data to the IBM cloud and this version of the project was ultimately scrapped.
Accomplishments that we're proud of
As two of our members were proficient using R, we were still able to create our original idea.
What we learned
We learned that sharing and collaborating on ideas was extremely important to ensure that all of us were on the same page in regard to the project. We also learned that it was okay to start of scratch again if a certain process was causing us to hit dead ends in the process of creating the application.
What's next for Who the Flu?
We hope this application could be applied to other diseases. Additionally, we would like to try and use the Watson's NLU function to incorporate machine learning into the program.