Sentiment PANalysis

example of terminal output

Inspiration

Eight hours deep into the hackathon, we were particularly lacking for inspiration. We had made a few proof of concept apps with the Myo armband, but didn't think we would be able to write a full app. That was when a Google Cloud Platform-promoting person walked by and told us all the things it was capable of. We were convinced when we saw the language sentiment analysis! Originally we wanted to use the translation function as well for education-based translation of sentiment, but we didn't have time to integrate it all.

What it does

It is a command line tool that allows searches of a number of different news sources and returns color-coded results of the 'sentiment' (ranging from -1 to 1) contained by using an API for that particular source (Twitter, HackerNews, CNN etc). You can even search a term on Google and see a sentiment analysis of the first page of Google results.

How we built it

We wrote a python app and used a number of APIs and dependencies.

Challenges we ran into

Setting up a python enviroment that was able to interface with Google Cloud Platform was a real challenge, as well as integrating the first API.

Accomplishments that we're proud of

Being able to tell sentiment by Google search and being able to select different news sources.

What we learned

Python may be easy to code in, but not to set up an environment for! We also learned about Google Cloud Platform, natural language processing, API integration,