Question

I'm trying to use ELKI to cluster a dataset of geolocations using OPTICS. I've understood that to extract the clusters, I need to use the OPTICSXi algorithm rather than OPTICS which computes just the clusters order.

I was wondering if you could give me more information on how the parameter xi works. I fixed this value at 0.009 but in a random way.

Was it helpful?

Solution

You can read up on the Xi parameter in

Mihael Ankerst, Markus M. Breunig, Hans-Peter Kriegel, Jörg Sander (1999).
OPTICS: Ordering Points To Identify the Clustering Structure
ACM SIGMOD international conference on Management of data. ACM Press. pp. 49–60.

it is a contrast parameter, the relative decrease in density. I usually try values such as 0.1 (= 10% drop in density). However, the exact drop in density to be expected heavily depends on your data set and parameters, for obvious reasons.

Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top