US 12,293,398 B2
Computerized system and method for applying transfer learning for generating a multi-variable based unified recommendation
Wei-Lun Su, Jiaoxi Township (TW); Pei-Ling Chen, Taipei (TW); Chia-Chuan Ho, Taipei (TW); and Tzu-Chiang Liou, New Taipei (TW)
Assigned to YAHOO ASSETS LLC, New York, NY (US)
Filed by YAHOO ASSETS LLC, Dulles, VA (US)
Filed on Mar. 23, 2020, as Appl. No. 16/826,545.
Prior Publication US 2021/0295410 A1, Sep. 23, 2021
Int. Cl. G06Q 30/0601 (2023.01); G06F 16/33 (2019.01); G06F 16/334 (2025.01); G06F 16/383 (2019.01); G06F 40/205 (2020.01); G06F 40/30 (2020.01); G06F 40/58 (2020.01)
CPC G06Q 30/0631 (2013.01) [G06F 16/3344 (2019.01); G06F 16/383 (2019.01); G06F 40/205 (2020.01); G06F 40/30 (2020.01); G06F 40/58 (2020.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising the steps of:
identifying, by a computing device, a request associated with a user;
parsing, via the computing device, said request, and based on said parsing, identifying information indicating words in said request and language information indicating a source language and a target language;
generating, via the computing device, multi-lingual word embeddings corresponding to a set of m words in both the source language and the target language based on the word information and the language information, the multi-lingual word embeddings comprising a first set of m word embeddings corresponding to the source language and a second set of m word embeddings corresponding to the target language, wherein m is a variable representing a numeric value;
identifying, via the computing device, product information for a set of products based on the request and the language information, the product information being at least in the source and target language;
generating, via the computing device, a multi-lingual product representation based on the product information and the multi-lingual word embeddings;
determining, via the computing device, a ranked set of results based on the request, said determination comprising applying a knowledge base corresponding to the source language to the multi-lingual product representation, and generating the ranked set of results; and
communicating, via the computing device, to the user, said ranked set of results.