Chung, S (2024) Potential issue identification for country risks in international construction projects from news articles. PhD thesis, Seoul National University, Republic of Korea.
Abstract
International construction projects often face unforeseeable changes in the business environments of host countries, known as country risks. Due to their uncontrollable nature, it is crucial for project participants to monitor the evolution of these risks for timely mitigation actions. In this regard, continuous monitoring of current issues is vital for identifying potential country risks. Consequently, news articles can serve as a valuable information source for practitioners gathering information about the host country issues.However, utilizing news articles to identify issues presents three primary challenges: (1) the absence of the method to determine the relevance of news articles to country risk; (2) the absence of the method to extract significant issues from news articles; and (3) the absence of the method to identify the relationship between qualitative issues and quantitative risks. To address the stated problems, this study set three research objectives: (1) to assess the method to determine the relevance of news articles to country risk; (2) to apply and validate the method to extract significant issues from news articles; and (3) to develop the method to identify the relationship between qualitative issues and quantitative risks.Literature review on the related fields revealed that the current body of knowledge lacks: (1) exploration of issue identification in construction; (2) utilization of news text data and unsupervised techniques; (3) consideration of qualitative factors in quantitative analysis of risk indicators; and (4) rigorous validation of news topic modeling methods using ground truth data. The research framework, which consists of three components, was proposed to fill the knowledge gaps and thereby realize the research objectives.The news categorization model was developed based on data collection and a literature review of country risk factors. The model showed an acceptable performance in automatically identifying risk-relevant news articles from the collection of news articles. Consequently, the model was applied to the BBC News Dataset to filter 435,697 news articles potentially related to country risk.Next, the semantic news topic extraction model was suggested and validated. This study successfully developed a topic extraction model to extract significant issues from the news text using text embedding and clustering methods. Moreover, an automatic topic labeling approach was proposed to enable the interpretation of the issues with minimal prior knowledge of the news dataset. The developed model was compared to the existing topic extraction model, showing a better quantitative performance in accurately identifying the issues and better quality in the identified issues.Finally, the methodology to identify the relationship between the issues and quantitative risk indicators was suggested. This study leveraged the method for explainable artificial intelligence and tested the methodology on three real-world risk indicators. Consequently, the issues potentially related to the unforeseeable fluctuations in these risk indicators were discussed based on the results of the proposed methodology.This study can contribute to the body of knowledge by filling the mentioned knowledge gaps by (1) showing that news articles can be used to identify issues regarding the country risks in international construction; and (2) suggesting a novel methodology to interpret the issues in regard of the unpredictable changes in quantitative risk indicators. Furthermore, this study can help practitioners in international construction projects by enabling them to monitor issues in the host country, ultimately contributing to early risk identification.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Thesis advisor: | Chi, S |
| Uncontrolled Keywords: | artificial intelligence; international construction; monitoring; quantitative analysis; risk identification |
| Index terms: | face, literature review, risk identification, quantitative analysis, dataset, body of knowledge, international construction project, evolution, exploration, methodology, validation, practitioner, mitigation, qualitative factor, clustering, modelling, artificial intelligence, monitoring, international construction, risk factor |
| Subjects: | environmental hazards, research methods, financial risk, environmental resource management, psychology, analytical methods, knowledge management, environmental science, strategic management, practitioner, data management, professional development, data analysis and analytics, data science, artificial intelligence, control systems |
| Topics: | Research Practice, Information Management, Cost Management, Roles and Professions, International Construction, Digital Applications, Organizational Design, Site Management, Engineering Principles, Sustainability |
| 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