4

There are many binary similarity measures (e.g. Jaccard, Sorensen, etc), each of them is sensitive to different properties of the compared sets. I would like to use the metric $S=\frac{N_{A\bigcap B}}{min(N_{A}; N_{B})}$, where $N_{A}$ is the count of set $A$. So basically I divide the size of the intersection by the size of the smaller set. I am sure I am not the first who found this out, and maybe it has some pretty name. Does anybody know?

ttnphns
  • 51,648
  • 40
  • 253
  • 462
deeenes
  • 153
  • 7

1 Answers1

6

Your measure seems to resolve to a distance defined by Simpson. See A Survey of Binary Similarity and Distance Measures page 44, equation 45.

whuber
  • 281,159
  • 54
  • 637
  • 1,101
spdrnl
  • 2,017
  • 8
  • 11
  • Thanks, this is exactly what I was looking for. Also this brief compendium is very nice, thanks for that too. – deeenes Jun 02 '15 at 16:40