Model operation support system and method

A model operation support system comprising:

• an interface device;
• a storage device in which target data and base data are stored;
• and
• a processor connected to the interface device and the storage device,
• wherein the processor
• calculates an inter-distribution distance between the target data and the base data for each of one or more second periods of an entire period that is a period between an end time of a first period and a relearning timing of a learning model of machine learning,
• the target data being data for each of the one or more second periods of the entire data,
• the entire data being data for the entire period and being data related to the learning model,
• the base data being data for the first period and being data related to creation of the learning model,
• determines whether the calculated inter-distribution distance is greater than or equal to a first threshold for each of the one or more second periods,
• specifies a time in accordance with each of the one or more second periods as a variation point when a result of the determination is true,
• determines one or more training data candidates respectively corresponding to one or more periods in the entire period from a part or all of the entire data, based on one or more variation points, and
• outputs the determined one or more training data candidates or meta-information about the one or more training data candidates via the interface device for relearning of the learning model.