3 Replies Latest reply: Oct 25, 2013 8:59 AM by Mark Kelly-Oracle RSS

    Cubic Clustering Criteria (CCC)

    944112
      Hello all,

      I'm new to Oracle and clustering and I was wondering does Oracle SQL Developer have anything around (CCC)? I'm trying to find the optimal number of clusters to use on a data set. I know I can change the # of clusters in the Algorithm settings, but is there anything in the interface to help with this decision to use the optimal number?

      Thanks all.
        • 1. Re: Cubic Clustering Criteria (CCC)
          Mark Kelly-Oracle
          Hi,
          Not currently but development is looking at this.
          Thanks, Mark
          • 2. Re: Cubic Clustering Criteria (CCC)
            StatMan

            Mark:

            I note your reply was in June/2012.

             

            Can you please tell us if there has been any progress on incorporating CCC?

            Thanks - MB

            • 3. Re: Cubic Clustering Criteria (CCC)
              Mark Kelly-Oracle

              HI MB,

              Check out the Oracle 12c release.

              There is a new Clustering algorithm called Expectation Maximization.

              See snippet of doc and link below.

              Thanks,Mark

               

              New clustering algorithm: Expectation Maximization

              In addition to enhanced k-Means and O-Cluster, Oracle Data Mining now supports Expectation Maximization, a probabilistic clustering algorithm that creates a density model of the data. The density model allows for an improved approach to combining data originating in different domains (for example, sales transactions and customer demographics, or structured data and text or other unstructured data).

              Because of the probabilistic nature of Expectation Maximization, its cluster assignment probabilities may be more reliable than those produced by k-Means or O-Cluster. Also, the Expectation Maximization algorithm automatically determines the optimal number of clusters needed to model the data.

               

              Changes in This Release for Oracle Data Mining Concepts