Inspiration
The lack of use of sentimental data in the world of DeFi. Where as Fundamental and Sentimental data is the driver of NFT price action. With Stackon, users can make calculated decisons about their digital assets using predictions from Stackon model.
What it does
Stackon predicts the future price of Digital assets/NFTs in the next 24 hours using both historical and sentimental price data.
How we built it
I used neural networks to train the model, then built an API for the model then used Hooks to POST and GET data from the model then I deployed on the almighty CRC Testnet.
Challenges we ran into
A whole lot, since this is a new dispensation of data usage I ran into data pre-processing issues with limited time because i became aware and joined the hackathon just a week before deadline. Even, the deployed site unexpectedly failed so i'm still working on debugging it as of this moment of submission (which is 10 minutes to dealine; lol.
Accomplishments that we're proud of
The neural network as an accuracy score of 97% and I'm proud that i was able to build this 1 week to the end of the hackathon as i wasn't aware earlier. Also, most API providers i needed to get historical data from delayed so i had to create a detailed data set from scratch using affiliated historical data.
What we learned
More sentimental price action theories and how machine & deep learning and transform the world of De-Fi.
What's next for Stackon
Well, Stackon will surely go live on the CRC Netwrok and will have more integrations and features and atlas will be available for all. Even I can't wait to see how Stackon Dr Market will transform the industry.
Log in or sign up for Devpost to join the conversation.