US 12,154,146 B2
Automated customer review matching
David Hoffman, Glendale, CA (US); Lusine Martirosyan, Glendale, CA (US); and Hemant Kashyap, Glendale, CA (US)
Assigned to ServiceTitan, Inc., Glendale, CA (US)
Filed by ServiceTitan, Inc., Glendale, CA (US)
Filed on Oct. 20, 2021, as Appl. No. 17/506,000.
Claims priority of provisional application 63/094,256, filed on Oct. 20, 2020.
Prior Publication US 2022/0122134 A1, Apr. 21, 2022
Int. Cl. G06Q 30/00 (2023.01); G06F 16/908 (2019.01); G06F 16/951 (2019.01); G06N 5/04 (2023.01); G06N 20/00 (2019.01); G06Q 30/0282 (2023.01)
CPC G06Q 30/0282 (2013.01) [G06F 16/908 (2019.01); G06F 16/951 (2019.01); G06N 5/04 (2013.01); G06N 20/00 (2019.01)] 11 Claims
OG exemplary drawing
 
1. A computer-implemented method for matching customer reviews with service providers, comprising:
sending, by at least one computing device to a client device associated with a first customer, a notification to be displayed via an application executing on the client device, wherein the notification comprises a request for the first customer to review at least one of a service or a good purchased by the first customer;
aggregating, by the at least one computing device, customer data associated with the first customer, wherein the customer data indicates at least a time that the first customer was directed to an online review service by the at least one computing device, wherein the online review service is communicatively coupled with the at least one computing device via an application programming interface (API);
retrieving, from the online review service, a plurality of customer reviews;
extracting review metadata for each of the plurality of customer reviews, wherein the review metadata for at least one customer review of the plurality of customer reviews comprises sentiment metadata, wherein the sentiment metadata indicates at least one of an emotional state or tone associated with the corresponding customer review;
matching, by a machine learning model, the first customer with at least one of the at least one customer review based at least on the customer data and the extracted review metadata for each of the plurality of customer reviews, wherein matching the first customer with at least one of the plurality of customer reviews comprises:
providing the customer data and the review metadata to the machine-learning model; and
receiving, from the machine-learning model, a statistical inference indicating a correlation between the first customer and the at least one of the plurality of customer reviews; and
updating the machine learning model based on validating the matching of the first customer with the at least one of the at least one customer review.