Figure 1
From: The languages of health in general practice electronic patient records: a Zipf’s law analysis
The Pareto plots for the Salford data showing the cumulative distribution function Pr(x) plotted as a function of frequency (x) for the subset of the Read codes used in the Salford corpus. a) diagnosis codes; b) procedure codes; c) medication codes. The data for diagnosis and procedure codes could be effectively modelled, at least in part of their range, by a power law (shown as the dotted lines in a and b). However, there was no range on which the medication data could be modelled by a power law, c).