Probably the simplest first approach, https://en.wikipedia.org/wiki/Euclidean_distance
But then, for music (I'm making a guess here), each dimension is not equally weighted, so if your first column represents the root key then you might want to give that more clout. There are more elaborate _" distance function "_ you can research. On Mon, Mar 05, 2012 at 01:37:27AM +0200, ronni montoya wrote: > Hi , i have a groups of messages each one containing an array, or a > groups of numbers in this way: > > [ 2, 5, 1, 3, 5 ( > > [ 2, 1, 5, 9, 1 ( > > [ 5, 3, 1, 8, 5 ( > > [ 7, 7, 3, 8, 2 ( > > > ... and so on > > > > > How can i order each one of these arrays (messages) based in its > similarities? i need that the most similar ones can be neighbours, so > i can trigger them in a ordered way > > how can i achieve this? > > > thanks > > > R. > > _______________________________________________ > Pd-list@iem.at mailing list > UNSUBSCRIBE and account-management -> > http://lists.puredata.info/listinfo/pd-list _______________________________________________ Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list