A visualization of how much Manchester United fans enjoyed each game from the first half of the 21/22 Premier League season.
Satisfaction ratings were gauged by running a sentiment analysis on replies to Manchester United's main Twitter account (@ManUtd). After every game, Man Utd's Twitter account posts a tweet with the final scoreline. The replies to those tweets were scraped analyzed for every game.
The code is designed to scrape and analyze replies to one tweet only. So the process was repeated 19 times for each game.
After extracting the replies, I used TextBlob to assign a polarity value to each reply. The polarity could be positive or negative. Then I took a sum of all the polarities to get a net value of total polarity for the match day. The sum was then adjusted for the number of replies (the number of replies fo each match day were different). The final value was calculated under the variable name: "Per".
Results for each match day were added to the excel file --> Sentiment Analysis.xlsx Further analyses were performed to assign a value of satisfaction levels by looking at which percentile they ranked in.
Results were visualzed on a bar chart using Excel and PowerPoint
- Sentiment Analysis.ipynb: Python file with the scraper AND the initial calculations to calculate the polarity of fan's reactions to each game
- Sentiment Analysis.xlsx: Excel file that contains calculations used to assign a satisfaction rating on a scale of 1-5.
- Man Utd satisfaction ratings Viz.png: Bar chart showing results