Scaling behaviour in the number of criminal acts committed by individuals

Abstract

Abstract

Understanding scale invariance in a minimal model of complex relaxation phenomena

“…We find it remarkable -even though it does not prove anything-that the statistical properties of the avalanches in our model are indistinguishable in practice from what has been reported for some detailed laboratory experiments. For instance, size corrections similar to the ones in (4) and (6) for τ and α, respectively, have been reported in avalanche experiments on rice piles [44]. Moreover, our values for the infinite case are strikingly close to the ones reported in magnetic experiments, e.g., τ ∞ = 1.77 (9) , α ∞ = 2.22 (8) and γ = 1.51(1) in Ref.…”

A random walk in the literature on criminality: A partial and critical view on some statistical analyses and modelling approaches

“…An investigation carried out on the Cambridge database considered the number of convictions n of each individual. The frequency f (n) with which each number of convictions was observed follows a power law as a function of the number of convictions, f = αn β [19]. This corroborates the general observation that a small number of offenders are responsible for a disproportionate share of total crime [42].…”

“…The most conspicuous characteristic of a power law distribution is that it has a fat tail at large values of n, but here the range of the latter is intrinsically restricted, since it cannot be larger than the total number of offenses committed by a single individual. Interestingly, the self-reported offenses of the Pittsburgh Youth Study follows a similar distribution [19].…”

“…The Cambridge Study [37] contains data of ≈ 400 young males in Great Britain over the period 1961-1981, selected because of the prior expectation of a high prevalence of convictions (about a quarter) among them. Interestingly, about 60% were never convicted during the 20 years spanned by the study ([64], see also [19]) meaning that the sample contains a relatively low number of offenders (about 160 individuals), a problem that also exists in the other databases. Besides the small number, there are other difficulties with longitudinal data.…”

Optimization of Criminal HotSpots Based on Underlying Crime Controlling Factors Using Geospatial Discriminative Pattern

“…As with most criminal activities, the counts of cells with same values in each grid map follow a power-law distribution [6]. A better way to fairly represent all the variables in one pattern is to categorize them and change the original values into categorized numbers.…”