A knowledge graph of 15,941 math states – proof as path-finding

This project constructs a knowledge graph from 15,941 mathematical statements, modeling proofs as paths between nodes to enable automated theorem proving via graph search. It is designed for researchers in AI and mathematics, offering a structured dataset to explore reasoning and proof discovery. Its interest lies in reframing complex logical deduction as a tractable path-finding problem, potentially advancing machine learning for formal mathematics.