5 ms·
GraphSense – A Scalable Cryptocurrency Analytics Platform
- atomical 8y agoI'm interested in Ethereum analysis because of the token economy. Specifically, I would like to see when a large amount of tokens are moved from a known exchange addresses.
- mewwts 8y agoSome friends of mine are doing this https://www.coinfi.com/signals https://www.coinfi.com/signals
- atomical 8y agoThanks for posting that. It looks like a service traders would appreciate. I'm looking for something open source though because my interest is intellectual and the "signals" type stuff is associated with pump and dump groups. There is also so much market manipulation going on from all sides that it's really impossible to take any signals seriously.
- hanchang 8y agoThanks for the feedback - cofounder of CoinFi here. We actually help fund the open source Ethereum ETL scripts created by our data engineer Evgeny here: https://github.com/blockchain-etl/ethereum-etl https://github.com/blockchain-etl/ethereum-etl In addition, we've contributed the ingest as a public data set to Google Cloud if you want to do your own analytics without running the architecture: https://cloud.google.com/blog/products/data-analytics/ethereum-bigquery-public-dataset-smart-contract-analytics https://cloud.google.com/blog/products/data-analytics/ethere... The compilation of exchange specific addresses is a long and tedious process, plus we hypothesize that exchanges would rather prefer to keep their addresses private, so we're not releasing that information though. However, one of our Signals products reads from the Ethereum blockchain in real-time and notifies whenever there are abnormal movements to/from exchanges. Unfortunately we're aware of the "pump and dump" association with signals - any suggestions for front footing that negative connotation and showing users that this is in fact different? One way we've thought about is to create a Signals Library that basically explains each signal we provide and details the behind what it represents, a rough overview of what triggers it, and the historical performance if you traded off of it. Any other suggestions would be greatly appreciated!
- eikxyz 8y agoWould love to play around with this, but getting this up and running seems daunting.
- ThomPete 8y agoLooks interesting. I would love to get involved on the design side. Just did some other work for a client creating a tool that allow you to analyze multiple cryptocurrencies and their underlying metrics. (Difficulty, close price, volatility etc) and are interested in exploring this field a little more. Here is small section of one of the directions we went: https://cl.ly/c997d4680e7c https://cl.ly/c997d4680e7c https://cl.ly/a41354620e82 https://cl.ly/a41354620e82 If you are looking for someone to help send me a PM (and yes for free I understand it's an open source project)
- philipodonnell 8y agoThat looks nice. Did you use a UI/CSS library?
- ThomPete 8y agoNo all hand done
- 273 8y agoYou should probably also check out https://cryvo.com/ https://cryvo.com/
- lmeyerov 8y agoThat's really fun! Funny enough, we put up a tutorial last week on looking at the silk road bitcoin / rogue DEA incident with Neo4j<>Graphistry (we do GPU viz): https://colab.research.google.com/drive/1TwYTDaBcMFxL6g5xLQzo44I9inELT29u https://colab.research.google.com/drive/1TwYTDaBcMFxL6g5xLQz... Would love to see if we can do even cleaner as GraphSense<>Graphistry!