Knowledge Extraction Method to Support Domain Integrated Design Methodology

Abstract

Nowadays, bio-inspiration has enhanced the creation of sustainable and innovative solutions to modern engineering problems. Nature could inspire mechanical engineers to develop innovative ideas as a great source for multifunctional and optimized designs. However, it is very difficult to extract desired design knowledge from primarily text-based databases and mainly focus on describing the biological system. The main objective of this study is to build a multi-label classification system to classify bio-inspired designs to support the Domain Integrated Design methodology. The proposed system integrates NLP and text mining with several machine learning models to learn and predict the functionalities of bio-inspired design. Various design functionalities were summarized based on the available resources from the AskNature database, then the main information extracted from the database, and they were labelled with corresponding multi-functionalities. Due to the high complexity of the multi-label classification system, multi-label classifiers were built based on different combinations of baseline classifiers and trained to classify selected AskNature pages. One case study was conducted to verify the impact of the proposed system. The results showed that the proposed system is feasible and would be a solution for classifying the bio-inspired design and functional basis knowledge extraction method to support DID methodology.