Inspiration
The idea behind this project was to create to be able to cluster collectibles in the Metaverse Space in the small neighborhoods. If Metaverse is the alternate universe, where your digital Avatars would be the resident, the interaction between these digital Avatars based on the personality type of humans who owned them interested me as a problem.
What it does
It simply classifies the Bored Ape Yacht Club NFTs into the OCEAN(openness, conscientiousness, extraversion, agreeableness, and neuroticism) framework of human personality types. These keywords are generated from using deep-learning CNN on images and the metadata provided by the opensea API and it's classified with captioned images with personality-defining words.
How I built it
I've used data generation by accessing the open sea API for BAYC and then ran CNN for the classification. I've utilized open AI to generate conversational models based on the data generated from the classification algorithm.
Challenges I ran into
I found data generation in this space a bit tricky and figuring out the features that'll make the model work was tricky as well.
Accomplishments that I'm proud of
I'm proud of creating clustering algorithms and parameter tuning capabilities that will enable Metaverse communities to form closer bonds with each other.
What I learned
I've learned a lot about the potential of this technology. Especially how effectively it could be used to build dApps in the Web3.
What's next for Bored Ape Trials
To generate the user interface and use a decentralized DESO chain to make fun contests for users to interact based on their collections.
Log in or sign up for Devpost to join the conversation.