US 11,756,093 B2
Systems and methods for generating personalized item descriptions
Ankur Anil Aher, Maharashtra (IN); and Susanto Sen, Karnataka (IN)
Assigned to Rovi Guides, Inc., San Jose, CA (US)
Filed by Rovi Guides, Inc., San Jose, CA (US)
Filed on Aug. 24, 2020, as Appl. No. 17/1,099.
Prior Publication US 2022/0058709 A1, Feb. 24, 2022
Int. Cl. G06Q 30/00 (2023.01); G06Q 30/0601 (2023.01); G06F 16/28 (2019.01); G06Q 10/10 (2023.01); G06Q 30/0201 (2023.01); G06Q 30/0282 (2023.01); G06F 40/186 (2020.01)
CPC G06Q 30/0623 (2013.01) [G06F 16/285 (2019.01); G06Q 10/10 (2013.01); G06Q 30/0201 (2013.01); G06Q 30/0282 (2013.01); G06Q 30/0631 (2013.01); G06Q 30/0641 (2013.01); G06F 40/186 (2020.01)] 17 Claims
OG exemplary drawing
 
1. A method for dynamically generating a product description personalized for a target user account, the method comprising:
obtaining, from one or more servers via a communication network, at least one of product description data associated with a product or product comment data associated with the product;
generating, based on at least one of the product description data or the product comment data, a product description template comprising fields to be populated;
storing the product description template in a memory;
identifying, by a recommender system, a first set of user accounts, each user account of the first set of user accounts having a similarity score with a target user that is above a similarity threshold;
obtaining, from one or more servers via the communication network, a first set of comment data associated with a product of interest, the first set of comment data originating from the first set of user accounts;
generating comment-based description data, based on comment data from the first set of comment data, wherein generating comment-based description data comprises:
analyzing comment data of the first set of comment data, wherein the comment data of the first set of comments are associated with the product of interest and comprise information items, and wherein the analyzing comprises identifying an information item in the comment data, and associating a tag to the information item, and wherein analyzing product data or comment data is performed using at least one of part-of-speech tagging, dependency parsing, or domain knowledge, and
wherein the first set of user accounts is identified by the recommender system using at least one of collaborative filtering, content-based filtering, and a knowledge-based system;
storing the comment-based description data in the memory; and
generating, for aural or visual presentation, a product description for the target user account, wherein generating the product description comprises:
retrieving the product description template from the memory;
retrieving the comment-based description data from the memory; and
populating the fields of the retrieved product description template based on the retrieved comment-based description data.