결정트리와 전문가시스템을 이용한 BIM설계 공간용도 지식화 방안 = Space Usage Knowledge Extraction from BIM Data by Decision Tree and Expert System

다국어 초록 (Multilingual Abstract)

This study is to intellectualize the existing building designs by extracting experience knowledge from BIM data not from human experts in order to secure expert knowledge in measuring the space layout quality. We extract expert rules from a decision tree generated by machine learning the space information in BIM data. This allows us to generate quantitative presentation of expert knowledge which has no concrete expression. Our study shows that it is possible to obtain rules only from data without any help of architects and overcome the challenge of securing their experience and knowledge in designs. This result suggests a way of measuring and evaluating the qualititive and logical space allocation which has been a challanging issue in the industry.

참고문헌 (Reference)

1 "Understanding Information Takeoff (ITO)"

2 Hillier, B., "The Social Logic of Space" Cambridge University Press 1984

3 Liu, H., "Rule Based Systems for Big Data: A Machine Learning Approach, Vol. 13" Springer 2015

4 Jin, C., "Exploring BIM Data by Graph-based Unsupervised Learning" ICPRAM 582-589, 2018

5 David Gunning, "Explainable Artificial Intelligence (XAI)"

6 Kim, H., "Developing Energy Efficient Building Design in Machine Learning"

7 "Decision Trees"

8 Krijnen, T., "Assessing Implicit Knowledge in BIM Models with Machine Learning, Modelling Behaviour" Springer 397-406, 2015

9 Bonsang Koo, "Applying novelty detection to identify model element to IFC class misclassifications on architectural and infrastructure Building Information Models" 한국CDE학회 5 (5): 391-400, 2018

10 Wang, L., "A Deep Learning Approach to the Classification of 3D Models under BIM Environment" 9 (9): 179-188, 2016