However, sometimes there may be no obvious knee, or there can be multiple knees, which makes it hard to decide In a clustering with minPts = k, we expect that core pints and border points' k-distance are within a certain range, while noise points can have much greater k-distance, thus we can observe a knee point in the k-distance plot. One heuristic approach is use ln(n), where n is the total number of points to be clustered. Unfortunately many cases we don't know the domain knowledge, especially after data is normalized. MinPts is best set by a domain expert who understands the data well. If your distance degenerates, no clustering algorithm will work.Īs Anony-Mousse explained, 'A low minPts means it will build more clusters from noise, so don't choose it too small.'. The first thing you need to check however - pretty much independent of whatever clustering algorithm you are going to use - is to make sure you have a useful distance function and appropriate data normalization. Naively, one can imagine OPTICS as doing all values of Epsilon at the same time, and putting the results in a cluster hierarchy. It's much nicer, but I believe it is a pain to implement in R, because it needs advanced data structures (ideally, a data index tree for acceleration and an updatable heap for the priority queue), and R is all about matrix operations. OPTICS is a successor to DBSCAN that does not need the epsilon parameter (except for performance reasons with index support, see Wikipedia).
You can try to do a knn distance histogram and choose a "knee" there, but there might be no visible one, or multiple. It again boils down to choosing whatever works on this data set and this minPts and this distance function and this normalization. A low minPts means it will build more clusters from noise, so don't choose it too small.įor epsilon, there are various aspects.
There is no general way of choosing minPts.
0 kommentar(er)
