Training data generation apparatus, model training apparatus, sample characteristic estimation apparatus, and chromatograph mass spectrometry apparatus

An apparatus for food freshness detection, comprising:

• a chromatograph mass spectrometry apparatus that obtains a plurality of samples of food, and measures a plurality of pieces of peak information for the plurality of samples by applying gas chromatography mass spectrometry (GCMS) to the plurality of samples, the plurality of pieces of peak information representing results of mass spectrometry of the plurality of samples;
• and
• a processor configured to:
• obtain the plurality of pieces of peak information and a plurality of pieces of characteristic information associated with the plurality of samples, wherein the plurality of pieces of characteristic information is assigned based on indicators of freshness for the plurality of samples,
• delete data on at least one peak from any of the plurality of pieces of peak information whose peak intensity is lower than a prescribed threshold value,
• further delete, when a coefficient of correlation of data between peaks among remaining peaks is equal to or larger than a prescribed value, data on one peak of the peaks from each of the plurality of pieces of peak information,
• configure a feature data map as a N×(M+1) table, wherein N rows correspond to the plurality of samples, M columns correspond to peak intensities at the remaining peaks, and a last column represents labels indicating the plurality of pieces of characteristic information, and
• output the peak intensities in the feature data map as training data to a model generator provided in or operatively connected to a chromatograph mass spectrometry apparatus;
• wherein the processor is further configured to:
• train a model using the training data;
• input, into the trained model, peak data corresponding to features of new set of samples;
• and
• determines, for the new set of samples, using the trained model, a classification indicating freshness of the new set of samples.