Title: A text mining-based approach for modelling technical knowledge evolution in patents

Authors: Geng Li; Zuhua Jiang; Xinyu Li

Addresses: Department of Industrial Engineering and Management, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai, 200240, China ' Department of Industrial Engineering and Management, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai, 200240, China ' Department of Industrial Engineering and Management, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai, 200240, China

Abstract: This paper proposes a method for portraying the evolution of patent technical knowledge (PTK) and provides enterprises with an understanding of the evolution of a specific field. The paper proposes a text mining-based knowledge evolution modelling method. Based on weighted factors, PTKs are obtained from the bibliographic title, abstract and claims (TAC) data on granted patents. Then, PTK clusters are captured by affinity propagation (AP) and expressed with the PTK topics extracted via latent dirichlet allocation (LDA). Furthermore, an association measurement for PTK is proposed, and a long-term evolutionary process for a specific field is revealed by an alluvial diagram. A case study of the new energy vehicles (NEV) industry is presented, and the results indicate that the proposed method can be used to automatically obtain technical knowledge from patent documents and to better measure and analyse the evolution of PTK relative to traditional evolution methods.

Keywords: text mining; knowledge evolution; PTK; patent technical knowledge; LDA; latent Dirichlet allocation.

DOI: 10.1504/IJTPM.2020.111499

International Journal of Technology, Policy and Management, 2020 Vol.20 No.4, pp.318 - 339

Received: 09 Aug 2019
Accepted: 11 Dec 2019
Published online: 30 Nov 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article