비정형 데이터 마이닝을 이용한 패션제품 개발 -업사이클링에 대한 소비자 의견을 중심으로-

Abstract

Given the increasing social interest in environmental issues, fashion design based on up-cycling is drawing attention as a sustainable, pro-environment alternative to new product creation. However, consumers in South Korea tend to perceive up-cycled products unfavorably . This study conducted big data analysis to understand consumer perceptions and desires regarding up-cycling to promote the development of consumer-oriented up-cycled fashion products. First, we collected and refined unstructured textual data on consumer opinions. Next, we conducted a survey to classify the refined data. A primary survey with a group of experts was used to remove unnecessary keywords, while a secondary survey administered to consumers helped to classify the resulting 54 keywords into five categories. We performed word cloud analysis on the classified unstructured data using NodeXL and R 3.6.1. For the purpose of product development, we used a combination of 34 keywords based on degree centrality from the “Design and Production Method”, “Material”, and “Item” categories (out of five total categories). Originality (scarcity), Temporality (historicity), and Eco-Friendliness (environmental ethics), which are attributes of up-cycling design, were reflected in the product, which was then developed with an emphasis on eco-friendly production processes and familiarity with products. Two fashion products were developed: an eco-bag produced by weaving cut waste yarn and waste clothing and a shirt produced by weaving waste leather. The study’s process and results propose a method for developing consumer-oriented up-cycled fashion products that can mitigate social problems and promote an eco-friendly product consumption culture while contributing to the use of big data in the fashion design industry.