You want to group items in your dataset by similarity and assign labels to each group. What is the preferred analytic method to use for this task?

K-means clustering

Naïve Bayesian Classifier

Logistic Regression

ANOVA

Refer to the Graphics that plots the Value of Within Sum of Squares for the number of clusters (k) chosen for a particular data set in a k-means clustering solution.

What is the optimum number of clusters that would recommend using for this dataset?

4

6

8

2

What are the outputs generated by a k-Means clustering Analysis?

The centroids of the discovered cluster and the assignment of each input datum to a cluster

The rules that associate each input datum to a class and the diameter of the discovered clusters

Within Sum of Squares for each discovered cluster and the overall cluster dispersion

Class association for each datum and class probabilities