An Information-Flow Control Model for Online Social Networks Based on User-Attribute Credibility and Connection-Strength Factors

Ehud Gudes, Nadav Voloch

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

7 Scopus citations

Abstract

During the last couple of years there have been many researches on Online Social Networks (OSN). The common manner of representing an OSN is by a user-based graph, where the vertices are different OSN users, and the edges are different interactions between these users, such as friendships, information-sharing instances, and other connection types. The question of whether a certain user is willing to share its information to other users, known and less known, is a question that occupies several researches in aspects of information security, sharing habits and information-flow models for OSN. While many approaches take into consideration the OSN graph edges as sharing-probability factors, here we present a novel approach, that also combines the vertices as well-defined attributed entities, that contain several properties, in which we seek a certain level of credibility based on the user’s attributes, such as number of total friends, age of user account, etc. The edges in our model represent the connection-strength of two users, by taking into consideration the attributes that represent their connection, such as number of mutual friend, friendship duration, etc. and the model also recognizes resemblance factors, meaning the number of similar user attributes. This approach optimizes the evaluation of users’ information-sharing willingness by deriving it from these attributes, thus creating an accurate flow-control graph that prevents information leakage from users to unwanted entities, such as adversaries or spammers. The novelty of the model is mainly its choice of integrated factors for user credibility and connection credibility, making it very useful for different OSN flow-control decisions and security permissions.

Original languageEnglish
Title of host publicationCyber Security Cryptography and Machine Learning - Second International Symposium, CSCML 2018, Proceedings
EditorsItai Dinur, Shlomi Dolev, Sachin Lodha
PublisherSpringer Verlag
Pages55-67
Number of pages13
ISBN (Print)9783319941462
DOIs
StatePublished - 1 Jan 2018
Event2nd International Symposium on Cyber Security Cryptography and Machine Learning, CSCML 2018 - Beer-Sheva, Israel
Duration: 21 Jun 201822 Jun 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10879 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference2nd International Symposium on Cyber Security Cryptography and Machine Learning, CSCML 2018
Country/TerritoryIsrael
CityBeer-Sheva
Period21/06/1822/06/18

Keywords

  • Information-flow networks control
  • Online Social Networks security
  • Trust-based security models

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science (all)

Fingerprint

Dive into the research topics of 'An Information-Flow Control Model for Online Social Networks Based on User-Attribute Credibility and Connection-Strength Factors'. Together they form a unique fingerprint.

Cite this