[Hidekazu Ikeno] Establishment of a novel spectrum analysis technology by machine learning and first-principles calculations

Research Director

Hidekazu Ikeno

Hidekazu Ikeno

Graduate School of Engineering, Osaka Prefecture University
Associate Professor
website

Outline

Electron spectroscopies are powerful experimental techniques to investigate the microscopic states which dominate macroscopic physical properties of materials. In this research, we construct the theoretical spectral database by accurate first-principle calculations, and establish a new technique for the quantitative fingerprint matching between the experimental and theoretical spectra by utilizing algorithms of the informatics. In addition, we develop the spectral analysis engine that can predict spectral shapes from local atomic arrangements by using machine learning techniques which enable us to make efficient analysis of large numbers of experimental spectra. The technologies developed in this research will lead to the sophistication of nanostructure analysis in advanced materials.

Quick Access

Program

  • CREST
  • PRESTO
  • ACT-I
  • ERATO
  • ACT-X
  • ACCEL
  • ALCA
  • RISTEX
  • AIP Network Lab
  • Global Activities
  • Diversity
  • SDGs
  • OSpolicy
  • Yuugu
  • Questions