Sentiment analysis comparing fake and real news.

During my statistics course at Pitzer College I worked on a poster for social justice issue using one or two tailed t-tests. I wanted to compare the sentiment, or how positive or negative text is, of fake and real news from a Kaggle Dataset containing the two. I used Jupyter notebook and Python NLTK to compare the sentiment, Pandas to process the Kaggle dataset, and seaborn to visualize the differences between fake and real news. I found that fake news was more likely to contain a negative sentiment compared to real news!

📰 View The Results!

The results from the project can be viewed here, the notebook here, and the dataset from here! Thank you so much!