Skip to main content
EPJ Data Science

Table 9 The AUC with and without individual dynamic feature for Theft prediction

From: Crime event prediction with dynamic features

Dynamic Features Brisbane New York City
[15–18) [21–24) [15–18) [21–24)
wod wd wod wd wod wd wod wd
Visitor Homogeneity 0.807 0.808 0.770 0.794 0.676 0.681 0.626 0.628
Visitor Entropy 0.807 0.818 0.770 0.785 0.676 0.680 0.626 0.629
Region Popularity 0.807 0.810 0.770 0.769 0.676 0.679 0.626 0.632
Visitor Ratio 0.807 0.823 0.770 0.788 0.676 0.676 0.626 0.628
Visitor Count 0.807 0.812 0.770 0.777 0.676 0.679 0.626 0.625
Observation Frequency 0.807 0.809 0.770 0.771 0.676 0.675 0.626 0.626
  1. wod: without dynamic features; wd: with dynamic features.
Back to article page