My research aims to understand the Earth's deep interior by observing elementary particles called neutrinos. Neutrino is a kind of elementary particle, which has no electric charge and is highly transmissive. Therefore, they can reach the detector while retaining the information about their generation. Observation of neutrinos produced by the decay of radioactive materials in the Earth's interior (geo-neutrinos) are expected to elucidate the composition and structure of the Earth's deep interior, as well as the mysteries of the formation of the Earth.
The KamLAND experiment, let by the Research Center for Neutrino Science, has advantages of a large 1000-ton liquid scintillator and a low-background environment to make the world's first geo-neutrino observations in 2005, and since pioneer a new field of neutrino geo-science. I investigate how many neutrinos arrive from the Earth's interior by analyzing the data. Of course, the data contains backgrounds, and it is important to estimate them precisely using simulations. Recently, I have been working on using machine learning to improve resolution and particle identification.