Measurement technology has been improved in many academic fields, and a large amount of high-dimensional observation data is being generated every day. In addition, as represented by the terms "big data" and "data science," the information science and mathematical infrastructure for handling data is flourishing. We believe that in order to innovatively improve and enhance the level of science and technology, it is essential to create an academic field that closely integrates information science, as represented by machine learning, and natural science, which consists of many disciplines, and we have named it high-dimensional data-driven science. In this lecture, I will give an overview of data-driven science and its key technology, sparse modeling.

　In recent years, the advent of next-generation sequencers (NGS) has revolutionized genome sequencing, spreading the research to various species. This presentation will introduce the correlation between polymorphisms and mutations in the canine genome and breed-specific temperament and constitution, genetic diseases, and cancer genomes using NGS, and look at the applications of canine genome research results.

　We will introduce our research on shape-memory piezoelectric actuators using ferroelectric polarization control, and small linear motors using resonance drive.