Research

May 9th, 2011

My research interests are in the structure, function and evolution of large-scale compelx networks in nature and society, with an emphasis on the information and communication networks, such as the Internet and the World Wide Web.

I am currenty working on the following research topics.

  • Internet
    • Measurement of AS relationships
    • Internet routing structure
    • Internet performance and robustness
  • Complex networks
    • Structural constraints
    • Network growth mechanisms
  • Network security
    • Immunology-inspired Internet defence
    • Secure communication for cloud computing
  • Social networks
    • Characterisation and modelling
    • Visualisation and prediction
    • Online social media networks
    • Gaming and rumour spreading on social networks
  • Transport networks
    • Tube, bus and bike networks

Research Grants
———————————————————————————————————–

Title: Statistical topological studies on large-scale complex communication networks
Funding bodies: The Royal Academy of Engineering and EPSRC
Role: Principle Investigator (Research Fellowship)
Value: £484,258
Years: 2007 – 2012
Reference: 10216/70 (R31980)

Title: SCALE (SMALL CHANGES LEAD TO LARGE EFFECTS): Changing Energy Costs in Transport and Location Policy
Funding body: EPSRC
Role: Co-Investigator
Value: £793,908
Years: 2009 – 2012
Reference: EP/G057757/1 (P10509)

Previous grants:

  • Royal Academy of Engineering Research Exchanges With China and India, 2010
  • UK Research Council Visiting Research Fellowship, 2009
  • Royal Society Travel Grant, 2008
  • Nuffield Foundation Newly Appointed Lecturer Award, 2005 – 2007

Highlights of previous research
———————————————————————————————————

Rich-club phenomenon

We discovered that the Internet exhibits a rich-club phenomenon where the best-connected nodes, rich nodes, are tightly interconnected with themselves forming a core group. This property is relevant to the network’s routing efficiency and robustness. We introduced the rich-club coefficient to quantify this porperty. Today the coefficient has been widely used as a key metric to characterise a wide range of networks.

Positive-Feedback Preference (PFP) model of the Internet

We introduced the positive-feedback preference (PFP) model, which precisely reproduces the largest set of characteristics of the Internet AS-level topology. The model is one of the most advanced Internet topology generators.

Modelling document networks

We reported that seemingly very different websites in fact share a topological invariant related to triangular structures [1]. We proposed the degree-similarity product (DSP) model which reproduces a number of important connectivity properties as well as content similarity properties for webpage and citation networks [2].

  1. Cheng, XQ and Ren, FX and Zhou, S and Hu, MB (2009) Triangular clustering in document networks.NEW J PHYS , 11 , Article 033019.
  2. Zhou, S and Cox, I and Petricek, V (2007) Characterising web site link structure. WSE 2007: NINTH IEEE INTERNATIONAL SYMPOSIUM ON WEB SITE EVOLUTION, PROCEEDINGS. (pp. 73 – 80). IEEE COMPUTER SOC.

Network visualisation – BOSAM

We introduced a novel method, called BOSAM, to visualise network topological structure without loosing any original connectivity information. It is effective in comparing networks and it reveals interesting network proporties not seen by statistical physics studies.

Comments are closed.