Expectation Maximisation. I think the typical classroom example is fitting k normal curves to data, and in that case, the best-fitting k normal kurves will give the best "Expectation". But the one thing that people don't like about this is that it's not guaranteed to find all of them.
They're O(n) iterations for a long time... On Jan 17, 6:51 am, mg <[EMAIL PROTECTED]> wrote: > Hi, > using E-M algorithm it is possible to divide dataset into K fuzzy > clusters. K is predefined input variable. I'm wondering how to detect > "best-fitting" K? Should I run fuzzy cluster many times, and choose > most probable K, or is there a better way? > > -- > Regards > Mariusz --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Algorithm Geeks" group. To post to this group, send email to algogeeks@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/algogeeks -~----------~----~----~----~------~----~------~--~---
