Inspiration

We had seen people from different regions of the world have vastly different opinions about climate change, and we were curious as to what factors impacted opinions, and which opinions gained popularity.

What it does

We wanted to find the factors that affected the sentiment of climate change tweets and how the sentiment impacted the interaction with the tweet.

How we built it

Using sentiment analysis we classified each tweet on a scale from (-1,1) to describe how negative or positive it was. Using a weather API, we were able to correlate geo-tagged tweets with the user’s current weather. With the collected data, we performed

Challenges we ran into

We ran into challenges getting data from the Twitter API. We had a limited number of requests and had difficulty getting data about the location of tweets since we did not have twitter premium.

Accomplishments that we're proud of

We were able to successfully compile a full dataset from Twitter with wanted attributes despite having a limit and difficulty with pulling the location.

What we learned

Our analysis showed that the temperature outside had a slightly negative correlation with the sentiment of tweets about climate change, and slightly positive tweets gained the most popularity. The most popular tweets tended to be slightly positive, on average, positive tweets got more attention than their positive counterparts. As the temperature increased, the sentiment values of the tweets decreased. This means that tweets were more negative as the weather got warmer.

What's next for Sentiment Analysis of Tweets

We hope to further explore different factors that impact a user's sentiment about climate change. We would be able to do this with more data points on the location of tweets. From this, we are curious about how political parties, geography, and the impact of climate change on the area affect the sentiment of the user.

Built With

Share this project:

Updates