Ya’ll know how the distribution of crime is closely correlated to the space in which it occurs. Nobody gets mugged on a busy high street. But ya wander down a darkened, deadend alley at ya peril. Right?
In recent years a number of eggheads have been a-speculatin’ and a-wondrin’ about the nature of urban space and whether its essence can be adequately captured mathematically. Can yer build an algorithm that will tell whether a neighbourhood is a crime hotspot without having to risk life and limb explorin’ it?
Now “O” Dima Volchenkov at the University Bielefeld in Germany has come up with a way to do it based on random walks. To work out how well a particular destination in the city is connected to the rest of city, she chooses another point at random and sets off walking at random from there. The number of steps it takes her to reach her destination is measure of its connectedness or segregation.
O Dima says that when you do this for all the nodes in a city you build up a map that clearly shows the most segregated areas.
She’s applied the idea to the canal system in Venice and the most segregated area turns out to coincide with an area called the Ghetto of Venice. That’s no coincidence, she says, because:
“In March, 1516 the Government of the Serenissima Repubblica issued special laws, and the first Ghetto of Europe was instituted. It was an area where Jews were forced to live and which they could not leave from sunset to dawn. The Ghetto existed for more than two and a half centuries, until Napoleon conquered Venice and finally opened and eliminated every gate (1797).”
Whether that corresponds to an area with high levels of crime today is another matter. But O Dima seems to have built herself a handy tool that could be hugely useful for city planners and homebuyers not to mention muggers and rapists.
Ref: arxiv.org/abs/0710.3021: Ghetto of Venice: Access to the Target Node and the Random Target Access Time