Inspiration

We wanted to build a platform that allowed us to quantify and visualize consumer opinions of companies in real-time.

What it does

'Opiniorated' uses sentiment analysis of real-time tweets to score a company's favor-ability by giving it a score from 1 to 5, with 5 being the highest score. Real-time opinion/sentiment analysis has the potential to give us another measure that could help us predict various trends 'ahead of time'. While there are a lot of numerical factors that are used to do the same thing, 'opiniorated' harnesses the power of Natural Language Processing to quantify various qualitative factors.

How we built it

We built 'opiniorated' using

  • the Natural Language Processing API from the Google Cloud Platform to perform Sentiment Analysis,
  • the Twitter Standard Search API to mine real-time/recent tweets
  • d3.js to make our visualization/UI

The backend was programmed primarily in python and the front-end in JavaScript, HTML and CSS.

Challenges we ran into

The biggest challenge we faced was figuring out how to do data/sentiment analysis in real time, since real-time analysis was harder to implement and also made us feel very uncertain about what to expect.

Accomplishments that we're proud of

We're proud of finishing this hack, and can't wait to improve and build upon it!

What we learned

We learned how to use the ML/NLP capabilities of the GCP, and also how to use the Twitter API.

What's next for OPINIORATED

We want to build 'opiniorated' into an even more comprehensible platform

  • by implementing a consolidated scoring system that takes into account data from various other platform such as news outlets, reddit and so on
  • finding trends between the 'opiniorated' generated score and the stock-market value fluctuations for a particular company. So, we believe that this could turn into a very involved quant-project!

Built With

Share this project:
×

Updates