1

I am a research scholar in data mining. I'm interested in C# implementation of K-Means clustering algorithm for mixed numeric and categorical data.

IgorS
  • 5,444
  • 11
  • 31
  • 43
Muthu Kumar
  • 19
  • 2
  • 3
    A general request for help and advice is not appropriate for StackExchange. You need to ask a more narrow and focused question that can be answered with specific responses. – MrMeritology Sep 17 '14 at 06:43
  • 2
    Some people are rather vocal in their conviction that k-means is not appropriate for categorical data, e.g., [here](http://stats.stackexchange.com/questions/115573/what-are-clustering-techniques-for-this-case/) or [here](http://stats.stackexchange.com/questions/49887/appropriate-cluster-method) or [here](http://stats.stackexchange.com/questions/31308/distance-function-for-categories-in-k-means) or [here](http://stats.stackexchange.com/questions/40613/why-dont-dummy-variables-have-the-continuous-adjacent-category-problem-in-clust). – Stephan Kolassa Sep 17 '14 at 12:27

1 Answers1

1

The solution is described here: K-Means clustering for mixed numeric and categorical data

C# implementation can be found in ALGLIB library, which I strongly recommend: http://www.alglib.net/translator/man/manual.csharp.html#gs_packages

Community
  • 1
IgorS
  • 5,444
  • 11
  • 31
  • 43
  • ALGLIB looks really interesting! I'm a Data Scientist working in a .NET shop, and am looking to learn C#. Could you tell me a little more about the library? – neuron Nov 21 '14 at 09:38
  • Hi, in couple words... Old, solid, rich enough, made by one author. For many languages. I have used it several times for Dot Net. – IgorS Nov 21 '14 at 12:36