| Title |
A Design Support Tool Driven by Design Recommendation based on User Preferences |
| Authors |
김성준(Kim, Seongjun) ; 김성아(Kim, Sung-Ah) |
| DOI |
https://doi.org/10.5659/JAIK.2020.36.12.11 |
| Keywords |
Interior Finishing Material; User Preference; Text Mining; Semantic Segmentation; Image Similarity |
| Abstract |
Aesthetic preference of residents for interior finishing materials is an important factor to consider in the process of determining finishing
materials, but the related research is limited. In addition, in the case of multi-family houses, designers cannot provide an alternative that
meets the preference of residents because designers create an interior design by guessing the preferences of the residents. This study proposes
a design support tool that collects and analyzes review data of Airbnb and provides designers with interior design cases preferred by
residents. The proposed design support tool extracts user preferences and material information about Airbnb's interior design case through text
mining and deep learning and recommends them to the designer. A case study was conducted on 858 rooms in Airbnb located in Seoul to
verify the proposed design support tool. The results indicate that it was possible to provide similar cases preferred by a large number of
users to the designer, and the designer could modify the design based on recommendations. |