Alpha release of CryptoSherlock — Ai crypto analysis service on GraphGrailAi Platform
Hi, i have a pleasure to announce Alpha release of CryptoSherlock!
- our new product on GraphGrailAi Platform, that understand what people say about crypto on the market.
https://cryptosherlock.graphgrail.com/
Our neural network understands and classify tweets from 0 to 1 by positive or negative category. Total Rating column is the aggregated sentiment of project or token.
EOS example: https://cryptosherlock.graphgrail.com/token/127
Stats: we collected 298 826 tweets, 73 ICO projects in rating, 448 tokens (postICO projects).
Tokens rating:
https://cryptosherlock.graphgrail.com/tokens
ICO rating:
https://cryptosherlock.graphgrail.com/projects
Roadmap:
- Aspect sentiment: why project got pos or neg.
- Crypto exchanges solution: Ai for scam posts detection (“send 1 eth to address … and get 2 eth”, etc) and similar problems.
- Smart categories: dive in project’s quality: technology opinions, product, updates, team, etc.
But the coolest thing is that we will give you an ability to create your own features (read: own App) with the help of Ai-designer we announced in Whitepaper. (soon)
This Alpha is quite simple, yes — it is up to you to make it better. Use tweets base and label tweets with categories you wish, i.e. better filter for negative tweets, is there are a prediction of a price in tweet or not — and many more.
You can test hypothesis fast.
GraphGrail Ai — is the world’s first Artificial Intelligence platform for Blockchain.
Official website: https://graphgrail.com
Medium: https://medium.com/graphgrailai
Telegram: https://t.me/GraphGrailAi_en
Twitter: https://twitter.com/GraphGrail
Etherscan: https://etherscan.io/token/0xf207ccfe6d1b90b0b7817ce2d5d262af1514e37b
Roadmap:
- Aspect sentiment: why project got pos or neg.
- Crypto exchanges solution: Ai for scam posts detection (“send 1 eth to address … and get 2 eth”, etc) and similar problems.
- Smart categories: dive in project’s quality: technology opinions, product, updates, team, etc.
But the coolest thing is that we will give you an ability to create your own features (read: own App) with the help of Ai-designer we announced in Whitepaper. (soon)
This Alpha is quite simple, yes — it is up to you to make it better. Use tweets base and label tweets with categories you wish, i.e. better filter for negative tweets, is there are a prediction of a price in tweet or not — and many more.
You can test hypothesis fast.