Find vegetation clusters

In the Project Window (Data tab), right-click on the vegetation data set. Choose Detect Boundary > Constrained Clustering from the pop-up menu (or from the main menu).

  1. General Tab

    1. Data: vegetation

    2. Clear the box next to "measure goodness of fit for multiple partitions" (or leave it cleared).

    3. Choose a name for the clusters dataset and the new boundary. Examples:

      1. 14 Veg Clusters

      2. 14 Veg Cluster Boundary

    4. Choose a target number of clusters: 14.

    5. Choose to detect boundaries using all variables, equal weights.

    6. Choose NOT to standardize data before detection (clear the checkbox). Data standardization is only necessary when you analyze multiple variables without a self-normalizing metric.

  2. Advanced Tab:

    1. Select Steinhaus as the metric type. The Steinhaus metric is a self-normalizing metric appropriate for frequency data (such as counts of individuals of different vegetation types).

    2. Retain linkage clustering.

    3. Set the connectedness parameter to 0.8.   This setting asks BoundarySeer to compare adjacent clusters based on the more dissimilar elements in each cluster before deciding whether to merge them. This is a more stringent setting than lower connectedness.

    4. Do not choose to subsample the data.

    5. Choose to cluster with the k-means refinement (checked as a default).

  3. Hit "OK".

  4. View Boundary: Choose to view the clusters and boundary in Map1.

  5. Interpret map and table output.

Next