Mining and Representing the Concept Space of Existing Ideas for Directed Ideation

Abstract

Design innovation projects often generate large numbers of design ideas from designers, users, and, increasingly, the crowd over the Internet. Such idea data are often used for selection and implementation but, in fact, can 1also be used as sources of inspiration for further idea generation. In particular, the elementary concepts that underlie the original ideas can be recombined to generate new ideas. But it is not a trivial task to retrieve concepts from raw lists of ideas and data sources in a manner that can stimulate or generate new ideas. A significant difficulty lies in the fact that idea data are often expressed in unstructured natural languages. This paper develops a methodology that uses natural language processing to extract key words as elementary concepts embedded in massive idea descriptions and represents the elementary concept space in a core–periphery structure to direct the recombination of elementary concepts into new ideas. We apply the methodology to mine and represent the concept space underlying massive crowdsourced ideas and use it to generate new ideas for future transportation system designs in a real public sector-sponsored project via humans and automated computer programs. Our analysis of the human and computer recombination processes and outcomes sheds light on future research directions for artificial intelligence in design ideation.

A Data-Driven Approach for Creative Concept Generation and Evaluation

“…The computational concept generation method, the Combinator, employed in the approach produces ideas in a random manner. Although randomness is an essential factor of creativity ( Carruthers, 2011), He et al (2019) has indicated that computer-based random combinations often lead to meaningless outcomes through an empirical study. Controlled computer-based combinations as well as human combinations have higher probabilities in producing novel and feasible outcomes (He et al, 2019).…”

Semantic Networks for Engineering Design: A Survey

“…It uses a technology space map as the knowledge base, which is constructed based on patent data and enables exploration of technical terms in technology domains. He et al (2019) constructed a semantic network of concepts based on their co-occurrences in a set of one thousand idea descriptions from an online crowdsourcing campaign via Mechanical Turk for reuse to inspire design ideation. concepTe (Acharya and Chakrabarti, 2020) is a decision-making support tool during the conceptual design stage offering aids in the designer's familiar domain, of which the knowledge base is grounded in the domain-agnostic SAPPhIRE model ontology.…”

Design Knowledge Representation With Technology Semantic Network

“…For example, network visualizations have been utilized to represent the whole technology space to support innovation and competitive intelligence (Luo et al, 2017(Luo et al, , 2018Sarica, Yan, et al, 2020), show the relations between components and subsystems to evalute designs (He and Luo, 2017;Pasqual and De Weck, 2012;Sosa et al, 2007) and inform design decisions (Kim and Kim, 2012;Sosa et al, 2007), discover the patterns of design activities (Alstott et al, 2017;Cash et al, 2014;Cash and Štorga, 2015), reveal the structure of design document repositories to guide retrievals (Fu et al, 2013;Luo et al, 2021), and represent mind maps (Camburn, Arlitt, et al, 2020;Camburn, He, et al, 2020) and concept networks (Chen et al, 2019;Chen and Krishnamurthy, 2020;Liu et al, 2020;Sarica et al, 2019Sarica et al, , 2021Shi et al, 2017;Song, Evans, et al, 2020;Souili et al, 2015) for design ideation uses. On the other hand, a few studies explored other visualization methods such as word-clouds (He, Camburn, Liu, et al, 2019; based on design description texts. Although one can rigorously read and study a design document or description to discover all the design-related entities and comprehend their relations, such a human process is tedious, labourintensive, and limited by the domain-specific knowledge of the reader.…”