Thousands of methods have been developed to predict stock performance, some taking up to dozens of parameters and use complicated calculations. However, we realize that the market is ultimately determined by demand and supply, which means the stock is worth as much as the investors value them. So we decided to try a different approach, to use sentimental analysis on stock market news to determine stock performance.
What it does
It finds news related the user-specified stocks from nasdaq.com, scrap only the news part from the website and perform sentimental analysis on every articles found. It then produces the average score on a scale of 0 to 1 to determine how analysts are valuing the specific stock (0 = negative, 0.5 = neutral, 1 = positive)
How we built it
We use python and Azure sentimental analysis to build the program.
Challenges we ran into
Figuring out how to scrape data and perform sentimental analysis, since no member of our team has any previous experience.
Accomplishments that we're proud of
We successfully retrieve the urls from nasdaq.com, scrape data from all of the news urls and perform sentimental analysis on each of them.
What we learned
Everything that we achieved today was also learned today since our team have minimal experience with Python and completely no experience when it comes to data scraping and sentimental analysis.
What's next for Stock Market Sentimental Analysis
It is a very interesting project and we'll definitely work on expanding it and improving it's accuracy. We are looking to define weighting for news articles that are from different dates, with more recent news being weighted more heavily. We will also expand our news sources from more than just nasdaq.com to include Alphaseeker, The Motley Fool and other reputable sources. Finally we are looking to build a user-friendly website that include data visualization, how closely the analysis is to the performance in the past and let users review each article we analyzed.
Log in or sign up for Devpost to join the conversation.