Abstract: Under the influence of rapid developments in artificial intelligence, machine learning, as an important component of it, has significantly enhanced the intelligence, informatization, and automation in many scientific research fields. Geotechnical engineering needs to accurately predict and analyze actual engineering, so efficient and accurate processing and analysis of huge data is the key technical requirement. Machine learning is becoming an important driving force for the development of geotechnical engineering because of its advantages in processing huge amounts of data. To fully understand the progress and effectiveness of machine learning in the field of geotechnical engineering, this paper reviews a large number of relevant literature and uses CiteSpace visualization analysis tools to organize them, deeply exploring the current research status and hot issues, and identifying existing challenges and development bottlenecks. Through literature review, it is found that extensive and in-depth research on machine learning has been carried out in the field of geotechnical engineering. However, the development content and research direction of this field show a certain degree of limitation; although new algorithms have injected new vitality into the field of machine learning, the application of the latest algorithm results in geotechnical engineering is not widespread. In view of this, it is urgent to find ways to resolve its limitations and strive to apply the latest achievements to practical engineering to further promote the intelligent level of geotechnical engineering.

Key words: machine learning, geotechnical engineering, intelligent, CiteSpace, visual analytics