Machine learning approach to the recognition of nanobubbles in graphene

Since the local and elastic strain induced by nanobubbles largely affects the transport properties of graphene, detecting and probing nanobubbles are important processes for research on electronic transport in graphene. In this study, we propose a means to recognize the presence of nanobubbles in graphene by analyzing electronic properties based on a machine learning approach. Our machine learning algorithm efficiently classifies the density of states spectra by the height and width of the nanobubbles, even in cases with a substantial magnitude of noise. The machine-learning-based analysis of electronic properties proposed in this study may introduce a changeover in the probing of nanobubbles from image-based detection to electrical-measurement-based recognition.