Skip to content

icebert04/sentiment-analysis-POC

Repository files navigation

Sentiment-analysis-POC

This is a proof-of-concept for a Sentiment Analysis Bot

NOTE This is still a work in progress so some functions won't properly work.


Description

The user can Type in a keyword for the bot to scrape Hacker News, and then it will rate the article titles that match the keyword and rate it with (positive, negative, or neutral).

one input next to two buttons, analyze, & clear results with some results at the bottom

Here we typed "Cannabis" as this was the only noun with a suitable subject for the bot to score.

As you can see the bot recognizes this as a negative sentence.

In the future, the next function to make is a search function that scrapes data after the search is used.


Here is the Jupyter Notebook where the bot runs

Two python files in a Jupyter Notebook App

The Bot uses the NLTK Vader for the Sentiment Analysis


Here is what I used to build it:

The Frontend

  • Next.js
  • React.js
  • Axios
  • Tailwind.css

The Backend

  • Python
  • CORS
  • BeautifulSoup
  • NLTK
  • Vader
  • Flask

Disclaimer This repo is for demonstration purposes only. It is not yet complete and it still needs to expand features.

Anyone is free to contribute to this dApp.

Hope you can learn a lot from this. Feel free to fork, and send your feedback on this Project

Thanks!