US 12,321,844 B2
AI-based keywork predictions for titles
Gilad Eliyahu Fuchs, Kfar-Saba (IL); and Yoni Acriche, Austin, TX (US)
Assigned to EBAY INC., San Jose, CA (US)
Filed by eBay Inc., San Jose, CA (US)
Filed on Feb. 5, 2020, as Appl. No. 16/782,945.
Prior Publication US 2021/0241073 A1, Aug. 5, 2021
Int. Cl. G06F 40/166 (2020.01); G06N 3/042 (2023.01); G06N 3/08 (2023.01)
CPC G06N 3/042 (2023.01) [G06F 40/166 (2020.01); G06N 3/08 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
receiving, by a network system, a proposed title for a publication from a user during a publication creation process, the proposed title comprising a plurality of tokens and the publication being associated with an item, each token of the plurality of tokens representing a non-stock word in the proposed title;
analyzing the proposed title by applying a prediction model to the proposed title, the prediction model being trained to identify an importance of a specific position of each token relative to other tokens in proposed titles in order to determine an importance of each token, whereby tokens that are the same token in a same position have a different importance from each other for different corresponding items or tokens that are the same token and in different positions for a same item have a different importance from each other, the analyzing including:
identifying the item or content of the publication that is being generated;
based on the item or content, identifying context and probabilities for tokens of titles of publications for similar items or content;
identifying a specific position of each token relative to other tokens in the proposed title;
examining a relationship between each token of the plurality of tokens and the specific position of each token relative to the other tokens in the proposed title in order to determine the importance of each token based on the item or content, a combination of the relationship and the specific position of each token relative to the other tokens affecting a probability assigned to each token; and
assigning, using a hardware processor of the network system, the probability to each token of the plurality of tokens in the proposed title from the identified probabilities based on the relationship and the specific position of each token in the proposed title relative to the other tokens in the title and the context for the tokens of the titles of publications for the similar items or content; and
causing presentation, by the network system, of a user interface that visually indicates the importance of each token of the plurality of tokens in the proposed title and presents the probability for each token.