The 7th prediction science seminar was held on 24th March delivered by Dr.Daisuke Matsuoka from the Advanced Visualization and Computation Research Group, JAMSTEC. In his talk, he explained how to extract important information and characteristic structures in large amounts of data for exploring potential phenomena, such as detecting tropical cyclones (TC) and their precursors using deep learning. There were three parts of this talk. Firstly, he introduced a deep learning approach for detecting tropical cyclones and their precursors using atmospheric simulation data (Matsuoka et al., 2018). Secondly, he discussed how to improve classification performance through data science skills (Matsuoka, 2021). Finally, he talked about using deep learning to extrapolate pre-TC classification.
For more information about the seminar series, please see the following page: