Title:
INFERENCE-ORIENTED APPLICATION TRAFFIC AND INDEX VECTORIZATION METHOD AND SYSTEM
Document Type and Number:
WIPO Patent Application WO/2022/206040
Kind Code:
A1
Abstract:
The present invention provides an inference-oriented application traffic and index vectorization method and system, comprising: step S1: quantizing known network traffic in a distributed system and original index information of related devices of the current traffic into tuples of the same type, to obtain a known network traffic information tuple and an index information tuple; step S2: quantizing unknown network traffic in the distributed system into tuples of the same type, to obtain an unknown network traffic information tuple; step S3: constructing a network on the basis of the known network traffic information tuple, the index information tuple and the unknown network traffic information tuple according to a time window and a priori association relationship, and sampling network description information by using a random walk method; and step S4, obtaining a network traffic vector and an index information vector by using an embedding algorithm on the basis of the network description information; and step S5: performing root cause analysis and index prediction according to the network traffic vector and the index information vector.
More Like This:
JPS63167955 | DIAGNOSIS SUPPORTING SYSTEM |
JPH07244591 | METHOD AND DEVICE FOR SEARCHING STATE SPACE |
Inventors:
ZHI FENGLIN (CN)
CAI XIAOHUA (CN)
CAI XIAOHUA (CN)
Application Number:
PCT/CN2021/138960
Publication Date:
October 06, 2022
Filing Date:
December 17, 2021
Export Citation:
Assignee:
SHANGHAI NETIS TECH CO LTD (CN)
International Classes:
G06N5/04
Foreign References:
CN113098735A | 2021-07-09 | |||
CN104468403A | 2015-03-25 | |||
CN111277570A | 2020-06-12 | |||
US9729571B1 | 2017-08-08 | |||
US8682812B1 | 2014-03-25 |
Attorney, Agent or Firm:
DUAN&DUAN LAW FIRM (CN)
Download PDF: