Dynamically identifying and evaluating key barriers to promoting prefabricated buildings: Text mining approach

Prefabricated buildings (PBs) have been widely perceived as environmentally friendly buildings and advocated globally. In many countries, the promotion rate of PBs remains slow, hindered by various barriers. With the change of external development environment under different periods, there is also the temporal dynamic evolution in barriers. Many studies have adopted some traditional methods to explore these barriers, but only focused on the research-conducting periods and ignored the changes of obstacles in different periods, which is the so-called dynamic nature of barriers. This would lead to the noncorrespondence between the previously studied barriers and the new situation of different periods. In this regard, this study aims to explore barriers dynamically and efficiently. Taking the online discussions of Chinese construction practitioners (CPs) as the sample, text mining techniques were performed to identify the barriers at different stages in the period from 2016 to 2021. Then, with a proposed sentiment scoring model, CPs' sentiment and attention level were combined to indirectly evaluate the impact degree of the corresponding barriers. In the end, the findings were verified in semistructured interviews. As a result, the changeable sentiments of CPs were tracked in the timeline, the evolution of key barriers at different stages and the internal driving force were revealed, and some new findings of barriers were identified. This paper not only makes a great theoretical contribution for PB studies, but also provides a research framework worthy of reference for researchers or government to regularly explore problems in the PB industry or other industries.