ABSTRACT
In recent years, machine learning technology has rapidly advanced and found widespread application in various domains. To further explore its potential in tunnel engineering, this study employs bibliometric data mining to systematically review machine learning research in the context of tunnel engineering and their interconnections. Using specific search terms, relevant information was extracted from the Web of Science Core database and analyzed with VOSviewer, a bibliometric mapping analysis software, to visualize the network of recently published literatures and investigate interactions between machine learning and tunnel engineering research through citation networks. The analysis identified four distinct categories of machine learning research in tunnel engineering. Moreover, an examination of the connection and keyword co-occurrence networks in these two fields revealed challenges such as limited samples of actual engineering data and the constrained applicability of conventional machine learning models in tunnel engineering. These findings provide valuable insights for prospective research in this area.
