DMBI Experiment no 4

Aim- Use WEKA to implement the following Clustering Algorithms-K- means, Agglomerative and Divisive.

kMeans

Number of iterations: 4

Within cluster sum of squared errors: 149.5177664581119

Missing values globally replaced with mean/mode

Cluster centroids:                                      Cluster#

Attribute                Full Data               0               1

                            (768)           (500)           (268)

==================================================================

preg                        3.8451           3.298          4.8657

plas                      120.8945          109.98        141.2575

pres                       69.1055          68.184         70.8246

skin                       20.5365          19.664         22.1642

insu                       79.7995          68.792        100.3358

mass                       31.9926         30.3042         35.1425

pedi                        0.4719          0.4297          0.5505

age                        33.2409           31.19         37.0672

class              tested_negative tested_negative tested_positive

Time taken to build model (full training data) : 0.02 seconds

=== Model and evaluation on training set ===

Clustered Instances

500. 65%)

    268 ( 35%)

Hierarchical clustering

Cluster 0

((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((1.0:0.16243,1.0:0.16243):0.01512,1.0:0.17755):

Cluster 1

((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((0.0:0.12784,0.0:0.12784):0.02126,((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((

Time taken to build model (full training data) : 1.37 seconds

=== Model and evaluation on training set ===

Clustered Instances

0      268 ( 35%)

1      500 ( 65%)