The clarity of disclosure in patents: An economic analysis using computational linguistics

This thesis aims to explore and demonstrate the use of computational linguistic analysis to measure the "readability" of patent documents. By using readability as a proxy for the extent of disclosure in patent documents, this thesis studies whether patent applicants may strategically choose the disclosure level for their patents, and how the disclosure level would affect the patent acquisitions and patent examination. This thesis introduces a new method to the quantitative economic analysis of patents, and generates research results with important implications for patent policy and practice