CPC G06F 40/30 (2020.01) [G06F 18/214 (2023.01); G06N 7/01 (2023.01); G10L 15/14 (2013.01); G10L 15/1822 (2013.01); G10L 15/22 (2013.01); G06F 40/216 (2020.01); G06F 40/279 (2020.01); G06F 40/295 (2020.01); G10L 15/16 (2013.01); G10L 25/30 (2013.01)] | 20 Claims |
1. A method comprising:
applying, by at least one processor, a natural language understanding (NLU) model to an input utterance in order to obtain initial slot probability distributions;
performing, by the at least one processor, a confidence calibration by applying a calibration probability distribution to the initial slot probability distributions in order to generate calibrated slot probability distributions, the calibration probability distribution having a higher number of dimensions than the initial slot probability distributions in order to reduce over-confidence of the NLU model;
identifying, by the at least one processor, uncertainties associated with words in the input utterance based on the calibrated slot probability distributions;
identifying, by the at least one processor, a new concept contained in the input utterance that is not recognized by the NLU model using one or more of the identified uncertainties that exceed a threshold value;
identifying a word or phrase from the input utterance associated with the new concept; and
labeling the identified word or phrase with a slot type of unknown.
|