This paper aims to classify and structure explicit objects and properties which can be utilized in KBimCode and develop a database of the Korean building Act. In order to conduct automated building code compliance checking, it is essential to translate permit requirements in Korean Building Act into a series of explicit and logic-based form for computers. As a basis of the first step in translation of Korean building permit requirements for the automated building code compliance, previous research shows how to analyze regulation sentences written in natural language. Based on this, 1200 target objects and their related properties were extracted. The objects were captured in noun phrases of regulation sentences and classified into a series of types according to their information. Through the process of removing the similar and ambiguous meaning of objects and properties, they were transformed into an explicit form. Then they were categorized into two types of generic building object and building regulation-specific objects. In particular, the latter that condition regulations require to check have are based on dynamically instantiable objects from the implementation perspective. The same process of object extraction and classification was conducted for identification of target object properties. As an application of the objects and properties database, the web-based database is suggested for the user-friendly interface to easily find and manage objects and properties and measure the stats and frequency of usage of them.
|Number of pages||8|
|Publication status||Published - 2017|
|Event||34th International Symposium on Automation and Robotics in Construction, ISARC 2017 - Taipei, Taiwan, Province of China|
Duration: 2017 Jun 28 → 2017 Jul 1
|Other||34th International Symposium on Automation and Robotics in Construction, ISARC 2017|
|Country/Territory||Taiwan, Province of China|
|Period||17/6/28 → 17/7/1|
Bibliographical noteFunding Information:
This research was supported by a grant (17AUDP-B127891-01) from the Architecture & Urban Development Research Program funded by the Ministry of Land, Infrastructure and Transport of the Korean government.
All Science Journal Classification (ASJC) codes
- Building and Construction
- Computer Vision and Pattern Recognition
- Artificial Intelligence