Real-Time Parameter Estimation for Modelling Malware Propagation on Business and Social Networks Within a Corporate Environment

EasyChair Preprint 3282, version 2

Abstract

Tackling malware that spreads through business and social networks is a big cybersecurity challenge for large organisations and enterprises. To address this problem, we propose a new real-time parameter estimation method for forecasting Trojan malware propagation in such an environment. We set up a novel framework to estimate the per-interaction transmission rate p and verify the results of the estimation through a combination of real and simulated data sets. We discuss the benefits of integrating interactions into malware propagation models and study the accuracy and performance of our estimator for the parameter p. We examine how this method enables us to incorporate early detection data into real-time forecasts and how we are thus able to model malware not yet seen before.

Keyphrases: Forecasting, Malware Propagation Model, Trojan malware, agent-based model, compartmental model, networks, parameter estimation, real-time, simulations, spreading agent, stochastic modelling, zeroday attack

@booklet{EasyChair:3282,
  author    = {Stephanie Kiss and Xiao-Si Wang and Jessica Welding},
  title     = {Real-Time Parameter Estimation for Modelling Malware Propagation on Business and Social Networks Within a Corporate Environment},
  howpublished = {EasyChair Preprint 3282},
  year      = {EasyChair, 2020}}