Method, system, and computer program product for local approximation of a predictive model

A system, comprising:

• at least one processor;
• and
• at least one non-transitory computer-readable medium comprising instructions to direct the at least one processor to:
• receive unclassified data associated with a plurality of unclassified data items;
• classify the unclassified data based on a first predictive model to generate classified data;
• select a first data item from the classified data;
• generate a plurality of generated data items associated with the first data item using a generative model by perturbing the first data item based on dimensions of the generative model;
• classify the plurality of generated data items based on the first predictive model to generate classified generated data;
• train a second predictive model with the classified generated data;
• and
• display probabilities associated with a plurality of possible classifications based on the second predictive model and the plurality of generated data items.

A computer program product for local approximation of a predictive model, the computer program product comprising at least one non-transitory computer-readable medium including one or more instructions that, when executed by at least one processor, cause the at least one processor to:

• receive unclassified data associated with a plurality of unclassified data items;
• classify the unclassified data based on a first predictive model to generate classified data;
• select a first data item from the classified data;
• generate a plurality of generated data items associated with the first data item using a generative model by perturbing the first data item based on dimensions of the generative model;
• classify the plurality of generated data items based on the first predictive model to generate classified generated data;
• train a second predictive model with the classified generated data;
• and
• display probabilities associated with a plurality of possible classifications based on the second predictive model and the plurality of generated data items.