	US 12,265,990 B2
	Method of detecting brand theft and and apparatus for performing the same
Ki Cheol Kim, Seongnam-si (KR); Dong Bin Seol, Suwon-si (KR); and Keun Tae Park, Seoul (KR)
Assigned to S2W INC., Seongnam-si (KR)
Filed by S2W INC., Seongnam-si (KR)
Filed on Dec. 14, 2023, as Appl. No. 18/540,268.
Claims priority of application No. 10-2022-0190080 (KR), filed on Dec. 30, 2022.
Prior Publication US 2024/0273571 A1, Aug. 15, 2024
Int. Cl. G06Q 30/02 (2023.01); G06F 16/951 (2019.01); G06Q 30/0241 (2023.01); G06V 30/10 (2022.01)

CPC G06Q 30/0248 (2013.01) [G06F 16/951 (2019.01); G06V 30/10 (2022.01)]

9 Claims

1. A method of detecting a brand theft, which is performed by an electronic device in e-commerce, the method comprising:

acquiring protected object data related to a brand to be protected, the protected object data including brand character data representing the brand, brand logo image data related to the brand, and content delivery network (CDN) keyword data indicating a source in which the brand logo image data is stored;

crawling an e-commerce web page by using a web crawling technique and acquiring crawled data as a result of crawling the e-commerce web page, wherein the crawling of the e-commerce web page comprises crawling the e-commerce web page by using a web crawling tool;

parsing the crawled data to obtain standardized data from the crawled data, wherein the obtaining of the standardized data comprises acquiring first data corresponding to a title or content of the web page, second data corresponding to an image of the web page, and third data related to a source of the image;

analyzing each of the first data, the second data, and the third data and detecting whether the protected object data is included in the crawled data for the web page; and

monitoring e-commerce websites based on a result of the detection,

wherein the detecting of whether the protected object data is included in the crawled data for the web page includes:

determining whether the first data includes a character corresponding to the brand character data of the protected object data,

determining whether character information in the image included in the second data corresponds to the brand character data of the protected object data,

determining whether the image in the second data corresponds to the brand log image data of the protected object data, and

comparing the third data with the CDN keyword data of the protected object data to determine whether the third data includes a character corresponding to the CDN keyword data of the protected object data,

wherein the determining of whether the first data includes the character corresponding to the brand character data of the protected object data includes:

primarily classifying the first data including the character corresponding to the brand character data of the protected object data through keyword matching between the first data and the brand character data of the protected object data, and

secondarily classifying whether the classified first data includes the character corresponding to the brand character data of the protected object data from the classified first data through a trained filtering model, and

wherein the trained filtering model classifies the character corresponding to the brand character data of the protected object data from the classified first data as a true positive value, and classifies a character literally corresponding to the brand character data of the protected object data from the classified first data as a false positive value.