Research
Development of an AI-Driven Imputation Model for Floor Area Ratio Missing Data Incorporating Parcel-Level Spatial, Temporal, and Land Use Variables
South Korea's building registry is missing a large share of its floor area ratio values. This paper develops an AI-driven imputation model that estimates them from parcel-level spatial, temporal, and land use variables.
About this paper
South Korea’s building registry is missing a large share of its floor area ratio values, and the gaps are not random. This paper develops an AI-driven imputation model that estimates those values from parcel-level spatial, temporal, and land use variables. Published in the Journal of the Architectural Institute of Korea, 41(12), 35-45, with Sunjae Lee and Bumjoon Kang. An interactive visualization accompanies the paper.