US 12,346,867 B1
Systems and methods for intelligent defaulting using machine learning
Mohammad Pirhooshyaran, Seattle, WA (US); Jiang Du, Kirkland, WA (US); Taeuk Jang, West Lafayette, IN (US); Jonathan Toner, Bellingham, WA (US); Feng Zhu, Bothell, WA (US); Tian Chen, Redmond, WA (US); and Ashish Satish Urankar, Bothell, WA (US)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on May 1, 2023, as Appl. No. 18/310,360.
Int. Cl. G06Q 10/08 (2024.01); G06N 20/00 (2019.01); G06Q 10/083 (2023.01)
CPC G06Q 10/083 (2013.01) [G06N 20/00 (2019.01)] 19 Claims
OG exemplary drawing
 
1. A method comprising:
receiving, by one or more machine learning models, first input data comprising one or more prior online shipment option selections performed via a user interface of a website by a first user;
determining, by a neural network classifier and using the first input data, a first score associated with a first online shipment option, a second score associated with a second online shipment option, and a third score associated with a third online shipment option, wherein the first score indicates a first probability that the first user will select the first online shipment option via the user interface of the website, the second score indicates a second probability that the first user will select the second online shipment option via the user interface of the website, and the third score indicates a third probability that the first user will select the third online shipment option via the user interface of the website, wherein the first online shipment option is associated with a first threshold, wherein the second online shipment option is associated with a second threshold, wherein the third online shipment option is associated with a third threshold, and wherein the first threshold, the second threshold, and the third threshold are different;
adjusting, using one or more processors and based on one or more non-user-specific factors, the first threshold to a fourth threshold, the second threshold to a fifth threshold, and the third threshold to a sixth threshold, wherein the fourth threshold, the fifth threshold, and the sixth threshold are different;
determining that the first score fails to satisfy the fourth threshold, the second score satisfies than the fifth threshold, and the third score satisfies the sixth threshold;
comparing, based on determining that the second score satisfies than the fifth threshold and the third score satisfies the sixth threshold, the second score and the third score, wherein the first online shipment option is disregarded based on the first score failing to satisfy the fourth threshold;
determining that the second score is greater than the third score;
automatically selecting, using one or more processors and based on determining that the second score is greater than the third score, the second online shipment option via the user interface of the website; and
presenting, via the user interface of the website, the first online shipment option and the second online shipment option with the first online shipment option being indicated as selected by default.