Swordfin

                                                                              ##Inspiration 

Urge of learning artificial intelligence and the idea of getting our hands into data science truly motivated us and last but not the least the price kept us moving.

                                                                             ##What it does

Our team Swordfin has come up with a model which predicts market movements while helping with decision making and providing visualizations. So basically our model collects all the information of stock prices and stock volumes right from 2012 to 2020 and help us to predict its values in the future with respect to the past, which would be very beneficial in the field of stock market.

                                                                           ##How we built it

Artificial intelligence plays a key role in our model which comprises of numpy, pandas, plotly , scipy , matplotlib , sklearn and tensorflow .Figma has nevertheless gave us the flexibility of creating a user interface design with intuitive , fast and responsive tools which is a platform to design our UI/UX model. We gathered all the information from Udemy and YouTube.

                                                                      ##Challenges we ran into

We being in our first year of engineering, are completely new to data science . Data Camp has been very informative as we got acquainted with artificial intelligence and Figma which are the gist of technology but during the process we got stuck with linear regression by vanilla RNN . Then we came up with solution which was Long Short Term Memory(LSTM). LSTM networks work better compared to vanilla RNN since they overcome vanishing gradient problems. In practice, RNN fail to establish long term dependencies.

                                                           ##Accomplishments that we're proud of

Our team is extremely proud of having a command on AI and Figma in our first year itself which has now paved our way in the field of data science .

                                                                ##What's next for Swordfin

Swordfin is all set to make a UI/ML -based sentiment analysis models, can be used to understand the sentiment from public tweets, which could be used as a factor while making a buy/sell decision of securities. For example the NLP model(sentiment detector) will interpret positive and negative comments and will convert it to label =1 and label =0.

Built With

• machine-learning
• numpy
• pandas
• python
• tensorflow