Patent-KG: Patent Knowledge Graph Extraction for Engineering Design

DS 116: Proceedings of the DESIGN2022 17th International Design Conference

Year: 2022
Editor: Mario Štorga, Stanko Škec, Tomislav Martinec, Dorian Marjanović
Author: Haoyu Zuo, Yuan Yin, Peter Childs
Series: DESIGN
Institution: Imperial College London, United Kingdom
Section: Design Information and Knowledge
Page(s): 821-830
DOI number: https://doi.org/10.1017/pds.2022.84
ISSN: 2732-527X (Online)

Abstract

This paper builds a patent-based knowledge graph, patent-KG, to represent the knowledge facts in patents for engineering design. The arising patent-KG approach proposes a new unsupervised mechanism to extract knowledge facts in a patent, by searching the attention graph in language models. The extracted entities are compared with other benchmarks in the criteria of recall rate. The result reaches the highest 0.8 recall rate in the standard list of mechanical engineering related technical terms, which means the highest coverage of engineering words.

Keywords: knowledge representations, artificial intelligence (AI), data-driven design