Figure 4

Are the recipe and ingredient preferences of different regions generated by the same process? The figure shows the CCDF of the ingredient and recipe popularity distributions of individual regions. The variable x on the x-axis corresponds to the number of visits, while the y-axis shows the number of recipes/ingredients that have received equal or more than x visits. One can see that the ingredient popularity distributions of different regions tend to be very similar, while for recipes we can observe slightly more regional variability. This observation is also apparent when comparing the standard deviations of the fitted α slopes (cf. Table 1) - higher for recipes (all) compared to ingredients (all).