Can Machines Understand Humans? Text Analytics App for Tweets

Humans are complex beings. Our feelings are often implied (e.g. Great!!!) instead of explicitly stated (e.g. I’m excited), not to mention social and cultural nuances like humour and sarcasm. If you’ve ever been confused and thought “what do they mean?”, you know exactly what I mean.

Machines, on the other hand, are structured and methodical. Concepts like sentiment analysis and topic modelling can be used to understand emotions and group a series of texts (i.e. Tweets, sentences, stories) into common topics to identify themes. This would be a time-consuming exercise for humans who would need to manually read through all the texts.

Tweet Analyzer App

Can we understand what people are saying and how they feel about a topic? I explored this question by creating an application using Streamlit with my good friend and full stack analyst, Domenic Fayad.

This self-serve application takes a keyword or hashtag which is used to get relevant tweets from the past week for analysis. Check out the full application here or see the preview below:

NHL Expansion Draft App

We also built a version of the original application to monitor sentiments and themes for the NHL expansion draft on July 21, 2021. See the NHL Expansion Draft Analyzer here!