US 11,704,500 B2
Techniques to add smart device information to machine learning for increased context
Alan Salimov, San Bruno, CA (US); Anish Khazane, San Francisco, CA (US); and Omar Florez Choque, Oakland, CA (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on Sep. 9, 2022, as Appl. No. 17/941,581.
Application 17/941,581 is a continuation of application No. 16/859,190, filed on Apr. 27, 2020, granted, now 11,468,241.
Application 16/859,190 is a continuation of application No. 16/388,838, filed on Apr. 18, 2019, granted, now 10,679,012, issued on Jun. 9, 2020.
Prior Publication US 2023/0021052 A1, Jan. 19, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 40/30 (2020.01); G06N 20/00 (2019.01); G06F 16/28 (2019.01); G06F 40/205 (2020.01)
CPC G06F 40/30 (2020.01) [G06F 16/283 (2019.01); G06F 40/205 (2020.01); G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. An apparatus to train a chatbot model, comprising:
a memory storing programming code; and
processing circuitry, coupled to the memory, wherein the processing circuitry has an input and an output, and is operable to execute the stored programming code that causes the processing circuitry to perform functions, including functions to:
receive first data and a desired result, the first data comprising non-dialog data from a chatbot training data structure, the desired result comprising a data structure based on the non-dialog data and the data type of the non-dialog data;
infer an inferred result, based on the first data via a chatbot model, the chatbot model trained with a standardized chain of values including individual relative probability values indicating a probability of certain data elements having a determined correspondence and respective relationship probability values indicating a probability of a relationship between respective data elements of the certain data elements having the determined correspondence, the standardized chain of values generated based on transformed data, the transformed data based on the non-dialog data;
determine a degree of accuracy for the chatbot model by comparison of the inferred result from the chatbot model to the desired result;
continue training of the chatbot model if the degree of accuracy is less than a threshold degree of accuracy; and
generate information to exchange with a user based on information input by the user if the degree of accuracy is greater than or equal to the threshold degree of accuracy.