Affiliations: [a] Researcher, Computational Intelligence and Design Informatics Lab., Wayne State University, Detroit, Michigan, USA | [b] Professor, Department of Industrial and Systems Engineering, Wayne State University, Detroit, Michigan, USA
Abstract: Recently, design is supported by crowdsourcing. The vast knowledge generated and contributed by crowds enhances a wide exploration of design ideas. However, limited amount of information called ‘scarcity’, non-guaranteed quality of contributions, and similar contributions made by unspecified participant groups make it difficult to systematically analyze and reuse the design concepts. This paper presents a formal analysis method for design concepts generated by crowdsourcing design activities. To enable the formal design concept analysis, design feature taxonomy is first developed by considering crowdsourcing design environment. In this taxonomy, various design features and participant evaluation features are constructed. Galois lattices-based formal concept analysis is employed as a design concept analysis method. The pivot power case is used to show that the presented method is applicable to a practical crowdsourcing design environment. Finally, precision and recall tests are conducted through a focus group interview and the results from the design analysis without and with the participant evaluation are compared.
