US 11,941,082 B2
Technologies for classifying feedback using machine learning models
Christian Dorn Anschuetz, Johnsburg, IL (US); Surekha Durvasula, Skokie, IL (US); Spencer Sharpe, Laramie, WY (US); and Kyle Michael Caulfield, Mount Prospect, IL (US)
Assigned to UL LLC, Northbrook, IL (US)
Filed by UL LLC, Northbrook, IL (US)
Filed on Apr. 12, 2019, as Appl. No. 16/383,202.
Prior Publication US 2020/0327372 A1, Oct. 15, 2020
Int. Cl. G06F 3/048 (2013.01); G06F 3/0482 (2013.01); G06F 18/21 (2023.01); G06F 18/214 (2023.01); G06F 18/2431 (2023.01); G06Q 30/0203 (2023.01)
CPC G06F 18/2148 (2023.01) [G06F 3/0482 (2013.01); G06F 18/2178 (2023.01); G06F 18/2431 (2023.01); G06Q 30/0203 (2013.01)] 12 Claims
OG exemplary drawing
 
1. A computer-implemented method for classifying product feedback, the method comprising:
generating, by a processor, a machine learning model using an initial dataset comprising a set of initial consumer feedback entries associated with a set of products, including classifying each initial consumer feedback entry of the set of initial consumer feedback entries as one of a first category or a second category;
accessing, by the processor, a consumer feedback entry associated with a product of the set of products, wherein the consumer feedback entry comprises an initial classification as one of the first category or the second category, and wherein the initial classification is either specified by a user who submitted the consumer feedback entry or by an external user who reviewed the consumer feedback entry;
applying the machine learning model to the consumer feedback entry to classify the consumer feedback entry as a subsequent classification;
determining, by the processor, that the subsequent classification is the other of the first category or the second category such that the initial classification specified by the user or the external user and the subsequent classification classified by the machine learning model are mismatched;
in response to determining that the subsequent classification is the other of the first category or the second category, displaying, in a user interface, an indication of the consumer feedback entry, the initial classification, and the subsequent classification;
receiving, via the user interface, (i) a selection to specify a final classification, of the first category and the second category, for the consumer feedback entry, and (ii) a comment entered into a comment box of the user interface and associated with the final classification; and
in response to receiving the selection and the comment:
classifying, by the processor, the consumer feedback entry with the final classification, and
updating the machine learning model to reflect the final classification of the consumer feedback entry.