	US 12,463,908 B2
	Traffic categorization method and device
Won Ki Hong, Pohang-si (KR); Jae Hyoung Yoo, Seoul (KR); and Ji Bum Hong, Ansan-si (KR)
Assigned to POSTECH RESEARCH AND BUSINESS DEVELOPMENT FOUNDATION, Pohang-si (KR)
Appl. No. 17/768,837
Filed by POSTECH RESEARCH AND BUSINESS DEVELOPMENT FOUNDATION, Pohang-si (KR)
PCT Filed Apr. 10, 2020, PCT No. PCT/KR2020/004905 § 371(c)(1), (2) Date Apr. 13, 2022, PCT Pub. No. WO2021/112344, PCT Pub. Date Jun. 10, 2021.
Claims priority of application No. 10-2019-0160530 (KR), filed on Dec. 5, 2019.
Prior Publication US 2024/0137323 A1, Apr. 25, 2024 Prior Publication US 2024/0236007 A9, Jul. 11, 2024
Int. Cl. H04L 47/2441 (2022.01); G06N 20/20 (2019.01); H04L 41/16 (2022.01); H04L 43/026 (2022.01)

CPC H04L 47/2441 (2013.01) [G06N 20/20 (2019.01); H04L 41/16 (2013.01); H04L 43/026 (2013.01)]

14 Claims

1. A method of classifying traffic, the method comprising:

receiving flow data including information on flow;

performing scaling on the flow data;

generating input data by removing redundant data from among the flow data on which the scaling is performed, on the basis of correlation between the flow data; and

classifying, by an apparatus for classifying traffic performing machine learning in advance, network traffic on the basis of the input data,

wherein the machine learning is performed by:

generating the input data on the basis of the flow data on which labeling is performed; and

extracting a learning sample on the basis of the input data,

wherein the extracting of the learning sample comprises:

acquiring a slope of each of the input data;

aligning the input data on a magnitude of the slope of each of the input data;

extracting a portion of the input data from the input data according to a preset ratio; and

extracting the learning sample on the basis of the input data on which sampling is performed.