A Preliminary Study on Attractiveness Analysis of Real Estate Floor Plans

摘要

Understanding correlations between real estate floor plans and their implicit qualities in a quantitative data format would be important in order to obtain deep neural networks that could extract features related to the attractiveness of apartments. This paper presents creation and analysis of the preliminary dataset based on the attractiveness of floor plans using six statements that explore qualitative and functional values of apartments such as a level of comfort, perceived size, novelty, and so on, using crowdsourcing. Results revealed several characteristics that matched with our general preferences for apartments.