Technology fitness landscape for design innovation: a deep neural embedding approach based on patent data

ABSTRACT

Technology is essential to innovation and economic prosperity. Understanding technological changes can guide innovators to find new directions of design innovation and thus make breakthroughs. In this work, we construct a technology fitness landscape via deep neural embeddings of patent data. The landscape consists of 1,757 technology domains and their respective improvement rates. In the landscape, we found a high hill related to information and communication technologies (ICT) and a vast low plain of the remaining domains. The landscape presents a bird's-eye view of the structure of the total technology space, providing a new way for innovators to interpret technology evolution with a biological analogy, and a biologically-inspired inference to the next innovation.

KEYWORDS:

• technology fitness landscape
• design innovation
• deep learning
• neural embedding
• patent

Disclosure statement

No potential conflict of interest was reported by the author(s).

Acknowledgment

This work was supported by the SUTD-MIT International Design Center, SUTD Data-Driven Innovation Laboratory (DDI, https://ddi.sutd.edu.sg/), and Shanghai Jiao Tong University under the grant of the National Natural Science Foundation of China (52035007, 51975360), Special Program for Innovation Method of the Ministry of Science and Technology, China (2018IM020100), and National Social Science Foundation of China (17ZDA020).

Notes

1 IPC denotes the International Patent Classification scheme, and UPC denotes the United States Patent Classification scheme.

2 We also performed a Non-negative Matrix Factorization topic modeling on the domains of the global peak to examine its theme. Results show that all of the identified topics are about information, electronics, and electrical technologies, as shown in Figure 7.