Dimensionality reduction and unsupervised clustering for EELS-SI

Jinseok Ryu; Hyeohn Kim; Ryeong Myeong Kim; Sungtae Kim; Jaeyeon Jo; Sangmin Lee; Ki Tae Nam; Young-Chang Joo; Gyu-Chul Yi; Jaejin Lee; Miyoung Kim

doi:10.1016/j.ultramic.2021.113314

Dimensionality reduction and unsupervised clustering for EELS-SI

Ultramicroscopy. 2021 Dec:231:113314. doi: 10.1016/j.ultramic.2021.113314. Epub 2021 May 13.

Authors

Affiliations

¹ Department of Materials Science & Engineering and Research Institute of Advanced Materials (RIAM), Seoul National University, 08826, Seoul, South Korea.
² Department of Physics & Astronomy, Seoul National University, 08826, Seoul, South Korea.
³ Department of Computer Science & Engineering, Seoul National University, 08826, Seoul, South Korea.
⁴ Department of Materials Science & Engineering and Research Institute of Advanced Materials (RIAM), Seoul National University, 08826, Seoul, South Korea. Electronic address: mkim@snu.ac.kr.

PMID: 34024663
DOI: 10.1016/j.ultramic.2021.113314

Abstract

A novel combination of machine learning algorithms is proposed for the differentiation of distinct spectra in a large electron energy loss spectroscopy spectrum image (EELS-SI) dataset. For clustering of the EEL spectra including similar fine structures in an efficient space, linear and nonlinear dimensionality reduction methods are used to project the EEL spectra onto a low-dimensional space. Then, a density-based clustering algorithm is applied to distinguish the meaningful data clusters. By applying this strategy to various experimental EELS-SI datasets, differentiation of several groups of EEL spectra representing specific fine structures was achieved. It is possible to investigate particular fine structures by averaging all of the spectra in each cluster. Also, the spatial distributions of each cluster in the scanning regions can be observed, which enables investigation of the locations of different fine structures in materials. This method does not require any prior knowledge, i.e., it is a data-driven analysis; therefore, it can be applied to any hyperspectral image.

Keywords: Cluster analysis; Dimensionality reduction; Electron energy loss spectroscopy; Spectrum imaging.