Design Knowledge Representation With Technology Semantic Network

Abstract

Engineers often need to discover and learn designs from unfamiliar domains for inspiration or other particular uses. However, the complexity of the technical design descriptions and the unfamiliarity to the domain make it hard for engineers to comprehend the function, behavior, and structure of a design. To help engineers quickly understand a complex technical design description new to them, one approach is to represent it as a network graph of the design-related entities and their relations as an abstract summary of the design. While graph or network visualizations are widely adopted in the engineering design literature, the challenge remains in retrieving the design entities and deriving their relations. In this paper, we propose a network mapping method that is powered by Technology Semantic Network (TechNet). Through a case study, we showcase how TechNet’s unique characteristic of being trained on a large technology-related data source advantages itself over common-sense knowledge bases, such as WordNet and ConceptNet, for design knowledge representation.

Natural language processing in-and-for design research

“…Sarica, Luo, and Wood (2020) obtain embeddings of over 4 million unique terms from the titles and abstracts across the US patent database. Using a web-based tool called TechNet, 31 they facilitate a search for these terms (Sarica et al 2021) and utilise the embeddings of these to construct a similarity network (Sarica and Luo 2021). To create an engineering alternative to WordNet, Jang, Jeong, and Yoon (2021) collect 34,823 automotive patents (IPC B60).…”

“…To form keyword summaries of patent search results, Noh, Jo, and Lee (2015) conduct an experimental study to find that it is best to extract 130 keywords from abstracts using TF-IDF and Boolean expression strategies. Sarica et al (2021) propose TechNet (Sarica, Luo, and Wood 2020) as a means to search and expand technical terms, which were extracted from the titles and abstracts in the patent database. To facilitate cross-domain term retrieval, Luo, Sarica, and Wood (2021) organise the output of TechNet into various domains that are associated with a knowledge distance measure.…”

Engineering Knowledge Graph From Patent Database

“…Then, we segregate these application numbers according to the year of application. For each batch comprising 10,000 patents, we crawl the GooglePatents 12 webpage using a web-crawler -BeautifulSoup 13 .…”

“…To meet the growing demands for engineering knowledge retrieval, representation, and concept generation-cum-evaluation, scholars have resorted to large scale, cross-domain, and commonsense knowledge graphs like ConceptNet and FreeBase [9]- [13] that may not provide facts that are technically dominant or relevant. Meanwhile, there is a lack of engineering-contextualized knowledge graph at the scale of ConceptNet and alike [14].…”

Data-Driven Design-by-Analogy: State of the Art and Future Directions

“…In recent years, the increasingly available design databases and rapidly advancing data science and artificial intelligence (AI) technologies have enabled new data-driven methods and tools that support DbA. For example, deep learning, knowledge graph, natural language understanding, and computer vision may support the analogy representation, retrieval, mapping, and evaluation processes [13,14,15,16,17,18,19]. To the best of our knowledge, there is no systematic review and structured analysis of the literature on data-driven DbA to date.…”