Inspiration
Reliable news is a crucial factor in investing decisions. However, accessing this news can often be a time-consuming and cumbersome process.
What it does
We developed a web application that provides investors with an end-to-end means of analyzing the news sentiment of all NASDAQ-listed companies.
How we built it
We scraped the investopedia glossary for all relevant financial terms using BeautifulSoup, and we scraped NASDAQ for all ~5000 stock ticker symbols and full company names. Afterwards, we built a real-time news lookup engine using Google News to provide relevant, up-to-date news regarding the said 5000 stock tickers. We incorporated this news lookup engine with a NLTK-based text summarizer, and a NLTK-based keyword text extractor. Finally, we developed a UI using Streamlit, an open-source app development platform, and hosted it on Heroku (https://frozen-sierra-39842.herokuapp.com/). Basically everything (scraping, backend, summarizer, keyword extractor) was built using python.
Challenges we ran into
We originally intended to scrape the NASDAQ glossary, which was more comprehensive than the investopedia glossary, but we ran into some issues. Also, the text summarization did not work on the Heroku app, but it worked locally.
Accomplishments that we're proud of
We built a platform that can provide fast and reliable news regarding all NASDAQ-listed companies
What we learned
We learned about how to use NLTK for Keyword detection.
What's next for Smart Tool for News Analysis
For future work, we can extend the functionality of this tool by allowing users to follow the stocks they like, get notifications on breaking news as well as doing more NLP analysis like sentiment analysis.
Log in or sign up for Devpost to join the conversation.