Kuo, Vincent (2019) Latent semantic analysis for knowledge management in construction. DSc(Tech) thesis, Aalto University, Finland.
Abstract
Knowledge management (KM) in the Architecture, Engineering and Construction (AEC) industry is hampered by the overload and fragmentation of heterogenous data, information and knowledge. This has led to an inability to identify, capture, manage and reuse the large amount of potential knowledge generated by AEC organizations, underpinned fundamentally by the tacit nature of AEC knowledge, as well as the lack of interoperability of information systems. We address these challenges by evaluating a computational machine learning technique known as Latent Semantic Analysis (LSA) in real world AEC contexts, with reference to KM and knowledge creation theory. Specifically, we investigate whether and how LSA can be used to infer insights from abundant unstructured natural language textual material, that would otherwise have required expert tacit knowledge to achieve. Our investigations are rooted in various AEC domain problem contexts, applicable across the lifecycle: 1) Mapping of Communities of Practice via modelling of personnel knowledge, skills and competences, 2) Management of constructability knowledge and inference of domain topical relationships, and 3) Extraction and processing of change issues from a BIM-based renovation project. For each, we articulate the respective domain problems through the lenses of KM and align LSA to address them. Our principle results demonstrate that valuable insights, traditionally dependant on domain-specific human reasoning, can indeed be inferred from readily accessible textual documentation (be it passively or actively generated) in each AEC context. As such, LSA can help to give quantitative substance to many conceptual KM conjectures and contribute to developing both practical and philosophical elements of KM in AEC. We demonstrate that LSA has a clear role in advancing domain-specific applications of information retrieval and case-based reasoning systems, ontological mediation between data models as a matter of interoperability, as well as reconciliation between the ...
| Item Type: | Thesis (Doctoral) |
|---|---|
| Thesis advisor: | Paavola, Juha |
| Uncontrolled Keywords: | civil engineering; knowledge management; latent semantic analysis; natural language processing; interoperability; community of practice; building information modelling |
| Index terms: | information system, information retrieval, modelling, machine learning, building information modelling, reasoning, mapping, data model, competence, renovation, lifecycle, knowledge creation, knowledge management, fragmentation, constructability, communities of practice, investigation, mediation, tacit knowledge, interoperability, documentation, case-based reasoning, personnel |
| Subjects: | systems and processes, business, construction integration, artificial intelligence, spatial and geospatial analysis, analytical methods, management, data management, project delivery, personnel development, renovation and retrofit, dispute resolution, data collection methods, knowledge management, data science, information systems, professional development, cognitive psychology |
| Topics: | Project Management, Human Resources, Design Practice, Digital Applications, Information Management, Engineering Principles, Organizational Design, Legal Issues, Research Practice |
| Descriptive scope: | 4 PCTA |
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