US 12,380,118 B2
Generating personalized user recommendations using word vectors
Ashish Gupta, San Jose, CA (US); and Ranjan Sinha, San Jose, CA (US)
Assigned to EBAY INC., San Jose, CA (US)
Filed by eBay Inc., San Jose, CA (US)
Filed on Oct. 14, 2021, as Appl. No. 17/501,855.
Application 17/501,855 is a continuation of application No. 15/295,318, filed on Oct. 17, 2016, granted, now 11,176,145.
Claims priority of provisional application 62/243,037, filed on Oct. 17, 2015.
Prior Publication US 2022/0035826 A1, Feb. 3, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/2457 (2019.01); G06F 16/248 (2019.01); G06F 16/31 (2019.01)
CPC G06F 16/24578 (2019.01) [G06F 16/24575 (2019.01); G06F 16/248 (2019.01); G06F 16/313 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
monitoring, via a device, user interaction with a first item listing of a network-based marketplace, the first item listing including a first plurality of words;
generating a word vector for a first word from the first plurality of words, the word vector indicating contextual relatedness represented by an edit distance between the first word and a second word from the first plurality of words;
assigning, by at least one processor, a score to the word vector based at least in part on the user interaction, the score indicating a likelihood of a search being performed using the first word together with the second word;
identifying a second item listing with which a user of the device has a prior interaction;
mapping a second plurality of words in the second item listing to the first word and the second word from the first plurality of words based on the word vector;
ranking the second plurality of words in the second item listing at least based on the score assigned to the word vector; and
based on the ranking, outputting, by the at least one processor, the second item listing on the device.