System and method for identifying the geographic origin of a fresh commodity

A system for detecting a geographical origin of a fresh commodity comprising:

• at software means for generating a plurality of neural network models, each having means for receiving as a training set a data set from a plurality of samples of a commodity of known origins analyzed for a plurality of elemental concentrations;
• means for presenting to each neural network model for classification a test data set from a plurality of samples of a commodity of unknown origins analyzed for the plurality of elemental concentrations;
• a bootstrap aggregating software routine for combining the results of the classifications for each sample in the test data set made by each neural network model;
• and
• a poller for determining from the bootstrap aggregating strategy a final classification of each sample in the test data set indicative of a geographical origin of the commodity.

A method for detecting a geographic origin of a fresh potato comprising the steps of:

• collecting a chemical composition profile of a sample of a fresh potato of unknown origin;
• comparing the profile with a data set comprising a plurality of chemical composition profiles collected from samples of fresh potatoes of known origin;
• and
• determining from the comparison a likely origin of the sample of unknown origin.

A system for analyzing a geographic origin of a fresh potato comprising:

• means for collecting a first trace element profile of a sample of fresh potato of unknown origin;
• a second trace element profile for a sample of a fresh potato of known origin;
• means for comparing the first trace element profile with the second trace element profile;
• and
• means for determining from the comparing means whether a geographic origin of the unknown sample is likely to be the same as the known origin.