Abstract
Directed contact networks (DCNs) are temporal networks that are useful for analyzing and modeling phenomena in transportation, communications, epidemiology and social networking. Specific sequences of contacts can underlie higher-level behaviors such as flows that aggregate contacts based on some notion of semantic and temporal proximity. We describe a simple inhomogeneous Markov model to infer flows and taint bounds associated with such higher-level behaviors, and also discuss how to aggregate contacts within DCNs and/or dynamically cluster their vertices. We provide examples of these constructions in the contexts of information transfers within computer and air transportation networks, thereby indicating how they can be used for data reduction and anomaly detection.
Highlights
Directed contact networks (DCNs) are temporal networks in which edges are directed (Holme 2015; Masuda and Lambiotte 2016)
Markov chain models for DCNs we show that a useful probabilistic model of temporally coherent paths can be constructed from T(C) alone
In “Data reduction and anomaly detection” section we show that even sophisticated malicious cyber-activity leading to so-called “low and slow” data exfiltration involves at least some system callscale directed contacts that can be readily detected through temporally coherent path identification and analysis
Summary
Directed contact networks (DCNs) are temporal networks in which edges are directed (Holme 2015; Masuda and Lambiotte 2016). We construct a natural inhomogeneous (that is, time varying) Markov model (Huntsman 2018a) for probabilistic modeling of potential flows that aggregate contacts based on a simple notion of spatiotemporal proximity. This model involves a single parameter, which in practice we set automatically with an intuitive heuristic. We emphasize that this Markov model is not statistical in the sense that it involves no learning, fitting, optimization or other estimation procedure Instead, it starts from a small number of symmetry and invariance requirements that any model with its goals ought to obey.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
