6

I have a question about DBSCAN.

The points here are classified as core points, border points or noise.

  • A point p is a core point if at least minPts points are within distance ε of it, and those points are said to be directly reachable from p. No points are directly reachable from a non-core point.

  • A point q is reachable from p if there is a path p1, ..., pn with p1 = p and pn = q, where each pi+1 is directly reachable from pi (so all the points on the path must be core points, with the possible exception of q).

  • All points not reachable from any other point are outlier

Consider the following image:

DBSCAN example

I don't understand, if there must be 4 points within a radius or a point needs to reach 4 another points, in order to be classified as a core point.
So consider the far right red point (just under the letter "e"). This is a core point. But within its radius are just 3 other points, not 4. So there should be 4 points within a radius including the one point itself?
There are just 3 other points reachable from that point.

DaL
  • 4,462
  • 3
  • 16
  • 27
user1170330
  • 209
  • 2
  • 7
  • 4
    The picture is the same as on [Wikipedia](https://en.wikipedia.org/wiki/DBSCAN). There, the figure caption says that MinPts = 3. Could your MinPts = 4 simply be an error? – Stephan Kolassa Feb 09 '16 at 13:08
  • @StephanKolassa The figure caption on the Wiki page says minPts = 4. They might have updated it. – VaM999 Mar 26 '21 at 21:58

1 Answers1

2

In a database, all points are equal.

  • The blue point has 1 point in its neighborhood - itself.
  • The yellow points have 2 points in their neighborhood each.
  • The red points have 4-5 points in their neighborhood each.

Note that the definitions don't say "minPts other points"; but "minPts points". You can't ignore the one point you already know (what if it has duplicates?)

Has QUIT--Anony-Mousse
  • 39,639
  • 7
  • 61
  • 96