Method for creating discriminator

Abstract

An object is to accurately detect peaks of various compositions, even in a case of unseparated peaks in which peaks of a plurality of compositions are superimposed. A computer acquires waveform data Dhaving a peak Pin a composition A measured by a data analysis device (S). Next, the computer acquires waveform data Dhaving a peak Pin a composition B measured by the data analysis device (S). Next, waveform data Dincluding unseparated peaks by superimposing the waveform data Dincluding the acquired peak Pand the waveform data Dincluding the acquired peak P(S) is generated. Next, the generated waveform data Dof the unseparated peaks is input as learning data, and the waveform data Dand Dcorresponding to the waveform data Dare input as training data in Step S. Next, machine learning is performed using the waveform data D, D, and D, and a learned model for estimating an accurate separation method of unseparated peaks is constructed based on the trained result (S).