Ciro Santilli @cirosantilli  40

We propose a methodology to automatically detect burn addresses. We manually classified

Our model identified 7,905 true burn addresses from a pool of 1,283,997,050 addresses with only 1,767 false positive.

Storing images for fun and posterity. We did not observe plain text messages encoded in Bech32 addresses. As our methodology is designed to identify burn addresses with a human-readable structure or easily identifiable patterns, we are not supposed to detect images encoded in burn addresses.