Dynamic Stakeholder-Associated Topic Modeling on Public Concerns in Megainfrastructure Projects: Case of Hong Kong-Zhuhai-Macao Bridge

Public concerns are essential for megainfrastructure project success, meanwhile profoundly influencing stakeholder performance. To understand the dynamics of public concerns and the project's duration is beneficial for decision makers and project stakeholders to improve their adaptability in an uncertain project environment. Therefore, this study proposes a dynamic stakeholder-associated topic modeling approach to identify, evaluate, and manage these public concerns by learning about the experience from large quantities of unstructured project documents. The approach is composed of three parts: the Topic Over Time text-mining model, the stakeholder relevance scoring system, and the managerial map. Based on the managerial map, the proposed "Mirror Z"and "Letter N"strategies guide the decision maker and project stakeholders in determining the management priority of public concerns considering the public popularity and stakeholder relevance in different project phases. A 16-year case of the Hong Kong-Zhuhai-Macao Bridge is used to validate the effectiveness of the proposed approach using 1,748 official project documents. The dynamics of 16 critical public concerns in the project are revealed and verified by reality. Furthermore, the management guides are summarized for decision makers, political groups, construction groups, and pressure groups. This study establishes a text-mining based method to provide useful guidelines for future megainfrastructure projects in view of stakeholder management.