An Approach of Knowledge Acquisition in Biologically Inspired Design

Abstract

Biologically inspired design (BID) is the rising discipline where biological phenomena is taken to inspire the designers to solve the engineering problems or challenges. However, state-of-the-art methodologies struggle with acquiring and transferring the BID analogies into the engineering design. This research presents an approach of knowledge acquisition in BID, which focus on the clustering of biotic knowledge cells. The proposed approach has two stages of clustering in biological domain. The first level is based on semantic information and the group average linkage (GAL) algorithm is employed. The second level is based on environmental information, and a new method named Hybrid Fuzzy C-means (HFCM) algorithm is proposed. The specific experiment results of the visual prosthesis design indicate that the proposed methodology is feasible. Compared to the conventional K-means and hierarchical clustering algorithm, the clustering efficiency and accuracy are both improved.