Entering the world of investing may seem scary, and to be frank, it is. After being bombarded by millions of numbers and data sets that seem "ever-so-gibberishy," many may feel unmotivated. Yet, anyone that is able to enter the realm of stocks knows how difficult it is to keep track of the market. Without any reliable measure of how a stock's value will change in the future, investors tear their hair our from missed opportunities, agonize over consistent losses, and dread over missed gains. With dv8, we seek to solve that problem.
What it does
Essentially, dv8 offers guidance throughout the investing process. Our sentiment analysis algorithm analyzes many news sources, scraping positive and negative words to develop an accurate score on a 0 (extremely negative) to 10 (extremely positive) scale. Along with that, we pull data from NASDAQ's Data-On-Demand API to draw an accurate trend of the stock's value over the past month. Based off of that, we predict a value for the next day's closing price. Our implementation of machine learning and TensorFlow provides risk analysis and value predictions of a given stock. By analyzing performance behavior and deviation, we provide a risk score from 0 (a safe bet) to 10 (extremely risky). Finally, we combine all of our statistics together to form the dv8 score from 0 (not worth investing) to 10 (extremely worth investing). Our goal with dv8 isn't necessarily to nurture dependence, but rather expedite the investing process and influencing positive returns.
Challenges we ran into
Our primary challenge when building this first model of dv8 was integrating our backend with our frontend. Essentially, the communication between the client and the server was challenging to figure out and develop. There's a lot of data to analyze and give back to the user, so it was key to have a seamless integration.
Accomplishments that we're proud of
We all have different skills and experience levels, and we were glad that we could come together to make this work, despite the many challenges that we ran into. Although our MVP of dv8 isn't as efficient as we wanted it to be, we see a bright future ahead for dv8, as we are all passionate about this idea and product.
What's next for dv8
We had a lot of ideas for dv8 coming in, but only some could make the cut during the course of the last 36 hours. In the future, we seek to provide more accurate results for our predicted price by implementing machine learning. Along with that, we plan to develop regression models for the complete set of data over the past five years of any given stock. Right now, dv8 is primarily for short-term returns. In the future, however, we want to be able to make accurate predictions that will potentially bring long-term returns. Overall, dv8 strives to achieve more accurate results, as well as developing a platform for investing similar to that of Robinhood or Acorn.