Better way for tracking viruses and rumours

VIRUSES SUCH as swine flu and also rumours on Facebook or trends on Twitter can now be tracked more efficiently because of an advanced mathematical method developed in Ireland.

Prof James Gleeson of the University of Limerick improved the accuracy of models that can track “spreading behaviours”, the movement of infections, e-mails or computer viruses across large networks.

“The mathematical problem is to connect individuals across a network topography,” he said yesterday.

Existing methods were somewhat “crude” but he improved on these and can now track the dispersal of entities across any kind of network with a high degree of accuracy.

His research paper, "High-accuracy approximation of binary-state dynamics on networks", is published in the current issue of Physical Review Letters.

The level of connectivity across a network provides information about how something will spread out from a source, said Prof Gleeson, who is co-director of the Mathematics Applications Consortium for Science and Industry (Macsi) at the university.

The more connectivity between individuals along a network, the faster the spread, he said.

“A disease will spread along the airline network. A rumour will spread along the Facebook network.”

His model can look at a network and then predict how big and how fast a rumour, for example, will spread.

It can also provide a warning about how a serious infection might spread outwards from a location on a network.

Knowing the spreading pattern likely to arise from, say, a swine flu outbreak could help inform health authorities about the most effective vaccination strategy, said Prof Gleeson, whose work is funded by Science Foundation Ireland.

“Better understanding how diseases spread can inform how vaccination should be rolled out and targeted at specific groups and so guide the response required by governments and healthcare.”

He said the World Health Organisation makes use of this kind of mathematical modelling to assess disease risks, but his new technique makes this much more accurate.

Computer viruses are also transmitted across networks, and infect other computers just like infectious agents that attack humans.

Prof Gleeson’s technique allows him to assess the spread of a YouTube video clip or gossip over Twitter – just like a cold spreads through a neighbourhood.