Wang, N (2022) Spoken dialogue system for information extraction from building information models using artificial intelligence. PhD thesis, University of Florida, USA.
Abstract
The construction industry is known for being information-intensive. Building Information Modeling (BIM) has attained increasing popularity as a tool to provide comprehensive information support for construction practitioners in the Architecture, Engineering, Construction, and Operation (AECO) industry. However, as more data is aggregated in BIM, it is increasingly challenging to extract building information from BIM models to support construction activities. Existing BIM Information Extraction (IE) methods are unable to provide a human-machine interactive conversation system for BIM practitioners to use natural language speech. With the development of Artificial Intelligence (AI) technologies, more opportunities arose for using speech and natural language tools in the AECO industry. AI technologies, such as a spoken dialogue system (SDS), enable a machine to converse with a human via natural language speech. “AI BIM”, which involves integrating conversational AI techniques with BIM, has become one of the future trends of BIM development. This dissertation aims to develop an intelligent Building Information Spoken Dialogue System (iBISDS) to facilitate the performance and efficiency of IE from building information models. The iBISDS enables construction practitioners to utilize natural language speech to extract building information from building information models and enables a machine to generate a spoken natural language response. Furthermore, the developed iBISDS can be used by users with limited BIM knowledge and experience. The developed iBISDS is based on Industry Foundation Classes (IFC) specification data format, a widely supported open standard in the AECO industry. This dissertation implements machine learning technologies for natural language processing (NLP) and ontologies to support IE from BIM models. The architecture and functionalities of the iBISDS are discussed in detail. Quantitative validation and three case studies are utilized to validate the developed system. The contribution of this dissertation will facilitate the adoption of conversational AI technologies in the AECO industry, and the architecture of iBISDS can be extended to other IE areas.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Thesis advisor: | Issa, R R and Anumba, C J |
| Uncontrolled Keywords: | artificial intelligence; building information modeling; case studies; construction activities; industry foundation classes; learning; machine learning; specification |
| Index terms: | construction activity, case study, efficiency, building information modelling, artificial intelligence, construction industry, dissertation, industry foundation classes, ontology, construction practitioner, functionality, intelligent building, specification, machine learning, practitioner, validation |
| Subjects: | practitioner, computational design, research dissemination and communication, construction operations, design features, data collection methods, artificial intelligence, contractual condition, professional development, education and knowledge transfer, design practice, performance management, industry analysis, information systems |
| Topics: | Research Practice, Information Management, Roles and Professions, Design Practice, Digital Applications, Contract Administration, Site Management, Quality Management |
| Descriptive scope: | 4 PCTE |
N.B. Descriptive scope is a count of how many of the five facets of empirical research are indicated by the words used in title, abstract and keywords. It is not intended as a judgement on the research; merely a count of the kind of word we would expect to indicate Phenomenon, Concepts, Theoretical framing, Empirical techniques, Analytical techniques. If all five are present, then a code of “5 PCTEA” will indicate this. If you feel the coding for this record is questionable, we welcome discussion around the terms we matched or the way we categorized them. The facet you would expect may not be coded, or a facet may be coded inappropriately. This can also bear on a larger question, of which facets should be treated as defining in construction management research. Please get in touch, and we will look at it. More details here