Today, due to changing needs and consumption habits, the need for energy is constantly increasing. Especially, the construction industry has a large share in global energy consumption and greenhouse gas emissions. In recent years, the industry has focused on especially the concept of performance because of its potential contribution to current problems in the built environment. With the development in computer technologies, the productive and creative potential of digital tools is opening new dimensions to the building industry every day. Developing artificial intelligence technologies and, specifically, machine learning techniques have started to play an increasingly important role in performance assessment. The purpose of this study is to present a review of the use of machine learning techniques in building energy performance assessment. For this purpose, the Web of Science database is used, and the articles published between 2010 and 2021 in the related literature are examined. A bibliometric analysis is conducted to determine research trends. With the results obtained, it is seen that the machine learning methods are frequently used in building energy performance areas especially in recent years, and it has been determined that these methods are very effective in performance assessment. The articles are mostly published in certain journals and studied by a few prominent countries. Certain building types and machine learning techniques are used more frequently in the studies. It is thought that the less studied methods can also be experienced in different areas, which may create various potentials. In general, it is clear that possible developments in this area will contribute positively to global energy consumption and greenhouse gas emissions.