Presentation Information
[18a-M_123-3]Automated Detection of Onset Energies in Photoelectron and Inverse Photoelectron Spectra via Piecewise Polynomial Fitting using Bayesian Information Criterion
〇(M2)Daichi Egami1, Yoshida Hiroyuki2 (1.Chiba Univ., 2.Chiba Univ.)
Keywords:
Bayesian Information Criterion,Signal-to-Noise Ratio,automated analysis
Toward the automated determination of ionization energy and electron
affinity from photoelectron and inverse photoelectron spectra, we have
developed a segmented polynomial fitting method. A major challenge for
full automation has been selecting the appropriate smoothing window
width, which depends on the signal-to-noise ratio of each spectrum. In
this study, we propose an algorithm that optimizes the smoothing window
using the Bayesian Information Criterion (BIC), balancing the residual
sum of squares and degrees of freedom. We demonstrate its high precision
and robustness through applications to experimental spectra with various
S/N ratio.
affinity from photoelectron and inverse photoelectron spectra, we have
developed a segmented polynomial fitting method. A major challenge for
full automation has been selecting the appropriate smoothing window
width, which depends on the signal-to-noise ratio of each spectrum. In
this study, we propose an algorithm that optimizes the smoothing window
using the Bayesian Information Criterion (BIC), balancing the residual
sum of squares and degrees of freedom. We demonstrate its high precision
and robustness through applications to experimental spectra with various
S/N ratio.