A structure-function knowledge extraction method for bio-inspired design

Chen Chen,Ye Tao,Yan Li,Qiyu Liu,Song Li,Zhong Tang
DOI: https://doi.org/10.1016/j.compind.2021.103402
IF: 10
2021-05-01
Computers in Industry
Abstract:Bio-inspired design is an innovative design method that uses analogies to biological systems to develop solutions for engineering problems. It offers a huge potential to produce innovative concepts. However, most designers face difficulties in extracting design knowledge from biological information. This paper proposes a structure-function knowledge extraction method, which combines dependency parsing and keyword extraction. The dependency parsing computes dependency relations between pairs of words in a sentence. Through the specific types of dependency relations, the potential knowledge including structures and functions can be identified. The keyword extraction is to obtain the terms that represent the subject of biological information. These terms target on filtering the identified knowledge, which ensures that the extracted structures and functions are all related to the subject. Among them, the structures and functions with association relationships are further combined into structure-function knowledge. The proposed method is preliminarily evaluated based on randomly selected test samples, and the result indicates that the method would be a promising alternative to extract structure-function knowledge. In addition, a case study illustrates the application of extracted structure-function knowledge in the improvement design of fan noise reduction. To verify the impact of the method on bio-inspired design, we conducted an experiment with three different groups. It demonstrates that the method can save the time of extracting design knowledge and provide the potentiality of generating novel ideas.
computer science, interdisciplinary applications
What problem does this paper attempt to address?