Ji Han

A Data-Driven Text Mining and Semantic Network Analysis for Design Information Retrieval

..................................................................................................................... 2 Declaration ................................................................................................................. 3 List of Publications ..................................................................................................... 4 Table of

Three driven approaches to combinational creativity: Problem-, similarity- and inspiration-driven

Creativity is a crucial element of design. The aim of this study is to investigate the driving forces behind combinational creativity. We propose three driven approaches to combinational creativity, problem-, similarity- and inspiration-driven, based on previous research projects on design process, strategy and cognition. A case study involving hundreds of practical products selected from winners of international design competitions has been conducted to evaluate the three approaches proposed. The results support the three driven approaches and indicate that they can be used independently as well as complementarily. The three approaches proposed in this study have provided an understanding of how combinational creativity functions in design. The approaches could be used as a set of creative idea generation methods for supporting designers in producing creative design ideas.

New directions in computational, combinational and structural creativity

This paper examines how new and creative relationships in datasets, not easily revealed by conventional information retrieval methods and technologies, can be identified using a mix of established and new methods. The authors present how the integration of computerised morphological analysis with new computational models, incorporating web crawler, data processing networking and data mining algorithms, can help facilitate the identification of new ideas. Boden’s concept of ‘Combinational Creativity’ indicates a structured process, which generates unfamiliar combinations of familiar concepts and constructs allowing creative styles of thought. This structured approach has been constrained by the resultant combinatorial explosion and the dearth of easily accessible computer software and supporting methodologies, to help identify viable new solutions. Feature-enhanced computerised morphological analysis provides a new structural support tool for creativity and innovation. Morphological analysis systematically structures and examines all the possible relationships in a multidimensional, highly complex, usually non-quantifiable problem space. Computerisation of the process now permits large number of configurations (millions) in the problem space to be majorly reduced (typicallyu2009>u200995%), identifying only internally consistent solutions. These solutions are likely to embrace configurations containing something, which has not previously been considered, thus increasing the probability of some form of technological or design breakthrough and hence truly creative.