Circular economy conceptualization using text mining analysis

Circular economy (CE) concept has received tremendous attention from both academia and industries. Every individual working in this field attempts to propose a definition to describe the key characteristics of CE, thus resulting in confusion about the global definition. This paper aims to provide a cumulative understanding of CE. For this purpose, a dataset of 172 definitions proposed from 2005 to date was gathered. This dataset was analyzed using text mining approach, typically topic modeling with Latent Dirichlet Allocation (LDA). The analysis resulted in 12 topics each containing 10 keywords that were prevalently used by scholars working in the CE field. The results were then interpreted in four categories created for the first time in this study. These categories include activities, strategies, aims and business models. Finally, a brand-new and comprehensive definition was proposed to summarize the findings. The proposed definition has extended the 4R framework (reduce, reuse, recover and recycling) by including a relevant design approach and incorporated environmental quality and social harmony considerations. Furthermore, this definition also highlighted the impacts of CE on the future generation. The proposed CE description is able to supply the CE community with a more comprehensive and deep understanding which could serve as a conceptual foundation for future work and implementation.