Analyzing Twitter sentiment and topics with Python and Streamlit
Data Science & Analytics

Analyzing Twitter sentiment and topics with Python and Streamlit

Data science app built using Python and Streamlit to run a text and sentiment analysis on Twitter data in real-time.

39 minutes, and 41 seconds

That was the average number of minutes that Twitter stole from me each day in the month of June.

And yes, I know what you’re probably thinking - is it even appropriate to summarize with the mean?!

Fake laugh GIF

There definitely were a few days that we can call… outliers. The median was closer to 30 minutes.

Building an app to analyze Twitter sentiment and topics

As I reflected on my mild Twitter addiction, I wondered what kind of sentiment & topics I was being subjected to.

Naturally, this evolved into a data science project with my good friend and data wizard Shannon Lo.

We decided to build an app that takes in a keyword and a number of tweets and outputs several text analytics insights that come from summarizing data from the Twitter API.

Tweet Analyzer Streamlit App

In a future post, I will be covering the logic of our app. For now, feel free to explore the app and check out the source code available in our github repo.

I’ve iframed the app below. You can check out the live app deployed on Streamlit here.

Next