Inspiration
The News Byte competition data.
What it does
It classifies stock in five classes: very bullish, slightly bullish, neutral, slightly bearish and very bearish and ranks stocks based on these classifications.
How we built it
It was built using Python's Pandas and Nltk libraries. Google Colab was the IDE used.
Challenges we ran into
Analyzing the large amount of data, working with NLP libraries and even just classifying the different articles. Also, I, initially, trained with the training data provided hoping that the test data would e provided. Unfortunately, this was not the case and I was forced to split the training data instead.
Accomplishments that we're proud of
The classification bit being achieved was the highlight of this competition for me.
What we learned
I learned how to work with textual data and learned quite a bit bout tokenization and lemmatization.
What's next for News Byte
There is a lot of insight one can gain from the data provided. Therefore, this is not the end for this competition.
Log in or sign up for Devpost to join the conversation.