Abstract
Combinational creativity is a common approach of creative idea generation. Invention arises from novel combinations of prior technologies. However, prior studies of creativity have suggested that overly novel combinations may be harmful to invention. There may be such a thing as “too much” or “too little” novelty that will influence future value of an invention. To figure out the “sweet spot” of novelty, 3.9 million patents are analyzed in terms of technology combination. It is found that inventions are more likely to have high value if they have moderate novelty in the center of their technology combination spectrums and high novelty in the breakthrough extreme of the combination spectrums, showing a novelty sweet spot in the central-extreme novelty matrix. Applying this theoretical understanding of combinational creativity to innovation practices, a case study is conducted for a commercialized patent-based product. This case study demonstrates the novelty evaluation method using the central-extreme novelty matrix. It also reveals the challenges and opportunities that originate from the novel combinations of the underlying technologies but only appear at the later stages of product development and manufacturing in the innovation process. Inspired by the theoretical understanding of combinational creativity, a directed ideation method is developed that extracts keywords embedded in massive idea descriptions and represents them with core-periphery word clouds for creative recombination. An ideation practice of future transportation system is conducted via humans and automated computer programs, to explore the potential of the ideation method. The analysis of the human and computer recombination processes suggests future research to develop computer-aided ideation systems that integrate both intelligent algorithms and human ideation. Additionally, several artifacts are developed as potential tools to facilitate combinational creativity-oriented design practices. The first is InNovelty, a patent assessment interface using the central-extreme novelty matrix. The second is a design card that presents core-periphery word cloud for directed ideation. The third is InnoStorm that prototypes a web-based interface for online brainstorming by automatic semantic networking. All these outcomes of theories, methods and tools on combinational creativity shed light on future research directions for data-driven innovation and creative artificial intelligence (AI) for design ideation.