Exit Menu

Patent knowledge design tool


Project description

Successful completion of the previous EPRSC-funded DASCIP project demonstrated the feasibility of achieving emerging design-prior art comparison through the application of engineering design working principles and ontology. The research framework developed in the original DASCIP project was mainly applied to the relationships between functional attributes and design structure for beverage can designs.

This project is funded by the EPSRC and is developing a design tool for automatically extracting knowledge that is contained in key sections of patents. This knowledge is then semi-automatically transformed into design requirements, functional attributes and design structure visualisations for use by designers in the design process.

Its broader application has been investigated in this project through an initial three-month secondment with Chiltern Railway. This resulted in a framework for design knowledge from patents that was broader than the original DASCIP project.

This framework is now being transformed into a design tool. In this design tool, automatic refers to the use of computer science tools to enable automated patent data extraction from patents on-line and/or in a database. Semi-automatic refers to the use of simple Natural Language Processing (NLP) steps that will be applied to enable visualisations. For example, transforming key terms automatically captured from the patent abstract into the production of visualisations that describe invention working principles.