US 12,254,871 B2
Machine learning system for customer utterance intent prediction
Abhilash Krishnankutty Nair, Ann Arbor, MI (US); Amaris Yuseon Sim, Ann Arbor, MI (US); Dayanand Narregudem, Ann Arbor, MI (US); Drew David Riassetto, Valley Park, MO (US); Logan Sommers Ahlstrom, Ann Arbor, MI (US); Nafiseh Saberian, Ann Arbor, MI (US); Stephen Filios, Canton, MI (US); and Ravindra Reddy Tappeta Venkata, Novi, MI (US)
Assigned to CHARLES SCHWAB & CO., INC., San Francisco, CA (US)
Filed by TD Ameritrade IP Company, Inc., San Francisco, CA (US)
Filed on Mar. 14, 2023, as Appl. No. 18/183,695.
Application 18/183,695 is a continuation of application No. 17/033,608, filed on Sep. 25, 2020, granted, now 11,626,108.
Prior Publication US 2023/0215426 A1, Jul. 6, 2023
Int. Cl. G10L 15/18 (2013.01); G06F 40/30 (2020.01); G06Q 30/02 (2023.01); G10L 15/14 (2006.01); G10L 15/16 (2006.01); G06F 16/9032 (2019.01)
CPC G10L 15/1822 (2013.01) [G06F 40/30 (2020.01); G06Q 30/0281 (2013.01); G10L 15/14 (2013.01); G10L 15/16 (2013.01); G06F 16/90332 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method of training neural network models to predict an intent of an utterance, the method comprising:
setting an encoder layer of a first neural network model to be trainable;
obtaining a subset of multi-word training utterances from among a first plurality of multi-word utterances, the first plurality of multi-word utterances including a plurality of multi-word utterances, from among a plurality of topic-tagged multi-word utterances, that are tagged with a first topic, from among a plurality of topics included in a topic set;
for each training utterance of the subset of multi-word training utterances,
inputting the training utterance into an input layer of the first neural network model to generate an embedding of the training utterance,
generating predicted intent values based on the embedding of the training utterance, the predicted intent values being a vector of generated probabilities, each of the generated probabilities being a probability that the training utterance corresponds to an intent of a plurality of intents,
determining a predicted intent of the training utterance based on the predicted intent values,
calculating an error value based on differences between the predicted intent values and training intent values, and
adjusting weights of a plurality of trainable layers of the first neural network model based on the calculated error values for each training utterance of the plurality of multi-word training utterances to reduce the calculated error values; and
training a second neural network model to predict an intent of a multi-word utterance based on second training data, the second training data corresponding to a second plurality of multi-word utterances, from among the plurality of topic-tagged multi-word utterances, that are tagged with a second topic from among the plurality of topics included in the topic set.