Computational Law for Patent Analytics: A Data-Driven Workflow on the KNIME Platform

Patent research involves processing large volumes of unstructured data, making search and analytics processes resource-intensive and expert-dependent. This paper addresses the challenge of automating these procedures at the intersection of computational law and the data-driven approach. An end-to-end data-driven pipeline for automating patent search and analytics on the KNIME platform was developed and tested. A formal model was created by adapting the CRISP-DM methodology for working with Rospatent open data, which confirmed the hypothesis about expanding the subject area of computational law to include patent landscape analysis tasks. The scientific and methodological contribution lies in presenting the workflow as a digital artifact – an executable protocol that formalizes the patent search and analytics procedure. It was demonstrated how this approach democratizes access to analytics, transforming the expert's role from routine search to strategic data interpretation.