Toward an Automated First Impression on Patent Claim Validity: Algorithmically Associating Claim Language with Specific Rules of Law

Abstract

Can an algorithm identify words that place patent claims at higher risk of invalidation? Today’s language processing software can hardly compete with a seasoned legal professional in assessing claim quality. It may, however, uncover patterns that indicate similarity of patent claim terms to others that have already been challenged in the courts. This methodology quantitatively approximates the invalidity risk a patent claim faces based on the similarity of its words to other claims adjudicated under a specific rule of law. In this way, software-based linguistic analysis according to the methodology presented here provides a starting point for efficiently and algorithmically generating a legal “first impression” on the text of a patent claim. This study explores potential correlation of keywords to patent eligibility, a legal doctrine restricting a patent’s monopoly power to innovations of particular types. This methodology explores the possibility of estimating risk for a particular claim, based on the presence of keywords commonly seen in claims previously adjudicated for validity under a specific rule of law. Such tools could efficiently reduce the scope of uncertainty for factors indicating patent quality and provide alternatives to costly litigation for patent value discovery.

Details

Publisher:
Stanford University Stanford, California
Citation(s):
  • Aashish R. Karkhanis and Jenna L. Parenti, Toward an Automated First Impression on Patent Claim Validity: Algorithmically Associating Claim Language with Specific Rules of Law, 19 Stanford Technology Law Review 196 (2016).
Related Organization(s):