Waveform-analyzing method, waveform-analyzing device and analyzing system

A waveform-analyzing method for analyzing a signal waveform which is a chromatogram or a spectrum with intensity on a vertical axis and time, wavelength, or mass to charge ratio on a horizontal axis, comprising:

• a model creation step for creating a trained model for locating a peak portion on the horizontal axis in an input waveform, by machine learning using a plurality of sets of reference waveforms in each of which a position of the peak portion is known;
• a region estimation step for dividing an analysis-target waveform into a plurality of partial waveforms on the horizontal axis, for determining whether or not each of the plurality of partial waveforms is a peak portion by using the trained model, and for estimating a plurality of different kinds of regions including a peak region and a non-peak region in the analysis-target waveform, based on a result of the determination;
• and
• a detection step for detecting an incorrect region estimation result by determining whether or not a sequence of the regions estimated in the region estimation step along the horizontal axis of the analysis-target waveform conforms to a predetermined regulation.