Leveraging Crowd Sourcing for Efficient Malicious Users Detection in Large – Scale Social Networks

0
714
Leveraging Crowd sourcing for Efficient Malicious Users Detection in Large-Scale Social Networks

Leveraging Crowd Sourcing for Efficient Malicious Users Detection in
Large – Scale Social Networks

Abstract

Leveraging Crowd Sourcing for Efficient Malicious Users Detection in Large-Scale Social Networks,The past few years have witnessed the dramatic popularity of large-scale social networks where malicious nodes detection is one of the fundamental problems. Most existing works focus on actively detecting malicious nodes by verifying signal correlation or behavior consistency. It may not work well in large-scale social networks since the number of users is extremely large and the difference between normal users and malicious users is inconspicuous. In this paper, we propose a novel approach that leverages the power of users to perform the detection task. 

Disadvantage

  • Different from the existing systems our work focuses on the scenario where the malicious user can not be easily detected by the system administrator or data correlation of nearly users. And we propose a novel crowdsourcing based approach totackle the malicious users detection problem.The malicious users in social networkshave a terrible impact on the network, in terms of degradingthe network’s performance, reducing the network’s efficiency,increasing the cost or even disabling the whole network.

Advantage

  • We introduce a novel, efficient, and effective approach,i.e., crowdsourcing, to detect malicious users in lagerscalesocial networks. Based on this, in order to encouragesufficient users to perform detecting tasks, weformulate the incentive mechanism design problem.
  • We solve the incentive mechanism design problem in twoscenarios: full information of users’ preferences and partialinformation of users’ preferences. In full informationscenario, we design the optimal incentive mechanism byordering users’ preferences. In partial information scenario,assuming that we only have statistical informationabout users’ preferences, we transform this problem to anoptimization problem and solve it by exploring the formof its solution.
  • We perform extensive simulations to illustrate the relationshipbetween the system’ total cost and factors, andvalidate our analysis.