How we used machine learning to classify one million Ethereum addresses

Aymen Jaffry
Jan 28 · 4 min read

Summary

Can we use machine learning to identify exchange-owned addresses on Ethereum?

Visualization of addresses on the first two principal components
This chart shows the correlation between different features and is used to select features.
Pearson correlation plot
This decision tree classifier is used to visualize the most determinant features in our model.

Conclusion


Join us

TRM Insights

We build software that transforms the way people use blockchain data. Today, our products are used by cryptonetworks, researchers, and financial institutions to solve their toughest problems.

Aymen Jaffry

Written by

Machine Learning Engineer

TRM Insights

We build software that transforms the way people use blockchain data. Today, our products are used by cryptonetworks, researchers, and financial institutions to solve their toughest problems.