Linguistic metrics for patent disclosure: Evidence from university versus corporate patents

Encouraging disclosure is important for the patent system, yet technical information in applications often inadequate. We use algorithms from computational linguistics to quantify effectiveness of applications. Relying on expectation that universities have more ability and incentive disclose their inventions than corporations, we analyze 64 linguistic measures applications, show university patents are readable by 0.4 SD a synthetic measure readability. Results robust controlling non-disclosure-related invention heterogeneity. The metrics evaluated panel “expert” student engineers further examined USPTO 112(a) – lack rejection. opens new research paths potentially facilitates improvement disclosure.