User Identity Linkage Method Based on User Online Habit

Liu, Yan (2020) User Identity Linkage Method Based on User Online Habit. EAI Endorsed Transactions on Security and Safety, 7 (26). e5. ISSN 2032-9393

[img]
Preview
Text
eai.22-6-2021.170240.pdf
Available under License Creative Commons Attribution No Derivatives.

Download (2MB) | Preview

Abstract

User Identity Linkage (UIL) across social networks refers to the recognition of the accounts belonging to the same individual among multiple social network platforms. Due to the network user's identities have the characteristics of various sources and real identity cannot be confirmed, it is very easy to become the main means of malicious user to carry out network attacks and spread rumors. User Identity Linkage not only can make the service provider to understand the user and thus to provide better service to the user, but also plays a significant role in improving the ability to find and track malicious users. For the credibility problem on the associated clues of user identification resulted from dynamic IP, shared Internet access and other factors, a user identity linkage method based on user online habit is proposed. This method assumes that the people use multiple network services crosswise when using the internet, converts the association analysis problem of user identification to the frequent pattern mining problem, and performs the optimization from three respective aspects: the online transaction database construction, the fast algorithm for mining frequent patterns and frequent co-occurrence identities consolidation. In order to improve the efficiency of frequent pattern mining, a parallelization of FPGrowth algorithm called MRFP-Growth algorithm is proposed to mine the user identifications of frequent co-occurrence quickly and efficiently. Experiments show that this method can associate multiple accounts of a user in network traffic with more than 85% accuracy in the scenario of dynamic variable IP address with only IP address and online time.

Item Type: Article
Uncontrolled Keywords: Network Traffic Analysis, User Identity Linkage, Frequent Pattern Mining, FP-Growth
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
QA75 Electronic computers. Computer science
T Technology > T Technology (General)
Depositing User: EAI Editor IV
Date Deposited: 09 Jul 2021 08:32
Last Modified: 09 Jul 2021 08:32
URI: https://eprints.eudl.eu/id/eprint/4407

Actions (login required)

View Item View Item