	US 12,019,994 B2
	Distance-based logit value for natural language processing
Ying Xu, Albion (AU); Poorya Zaremoodi, Melbourne (AU); Thanh Tien Vu, Herston (AU); Cong Duy Vu Hoang, Wantirna South (AU); Vladislav Blinov, Melbourne (AU); Yu-Heng Hong, Carlton (AU); Yakupitiyage Don Thanuja Samodhye Dharmasiri, Melbourne (AU); Vishal Vishnoi, Redwood City, CA (US); Elias Luqman Jalaluddin, Seattle, WA (US); Manish Parekh, San Jose, CA (US); Thanh Long Duong, Melbourne (AU); and Mark Edward Johnson, Castle Cove (AU)
Assigned to Oracle International Corporation, Redwood Shores, CA (US)
Filed by Oracle International Corporation, Redwood Shores, CA (US)
Filed on Nov. 30, 2021, as Appl. No. 17/456,916.
Claims priority of provisional application 63/119,459, filed on Nov. 30, 2020.
Prior Publication US 2022/0171947 A1, Jun. 2, 2022
Int. Cl. G06F 17/18 (2006.01); G06F 40/35 (2020.01); G06N 20/00 (2019.01); H04L 51/02 (2022.01); G06F 40/205 (2020.01); G06F 40/253 (2020.01)

CPC G06F 40/35 (2020.01) [G06N 20/00 (2019.01); H04L 51/02 (2013.01); G06F 40/205 (2020.01); G06F 40/253 (2020.01)]

24 Claims

1. A method comprising:

receiving, by a chatbot system, an utterance generated by a user interacting with the chatbot system;

inputting, by the chatbot system, the utterance into a machine-learning model comprising a set of binary classifiers, wherein each binary classifier of the set of binary classifiers: (i) is configured to estimate a probability that the utterance corresponds to a class of a set of classes; (ii) is associated with a modified logit function that transforms the probability for the class into a real number, wherein the modified logit function is a logarithm of odds corresponding to the probability for the class, the logarithm of odds being determined based on a distance measured between the probability for the class and a centroid of a distribution associated with the class;

generating, by the machine-learning model, a set of distance-based logit values for the utterance, wherein each distance-based logit value of the set of distance-based logit values is generated by:

determining, by a respective binary classifier of the set of binary classifiers, a respective probability that the utterance corresponds to a class associated with the respective binary classifier; and

mapping, by the respective binary classifier and based on the modified logit function, the respective probability to the distance-based logit value, wherein the mapping includes using a respective distance measured between the respective probability and a centroid of a distribution associated with the class associated with the respective binary classifier;

applying, by the machine-learning model, an enhanced activation function to the set of distance-based logit values to generate a predicted output, wherein the predicted output identifies a normalized probability predictive of whether the utterance corresponds to a particular class of the set of classes within a probability distribution, and wherein the enhanced activation function includes a learned parameter for normalizing an initial output of the enhanced activation function to determine the normalized probability; and

classifying, by the chatbot system and based on the predicted output, the utterance as being associated with the particular class.