Aim: Demonstration of clustering rule process on iris.arff using simple k means algorithm.
The experiment illustrates the use of simple k-means clustering weka explorer.
The sample data set use for this experiment is iris data in arff format.
Steps involved in experiment:
Step 1:create iris.arff file
@relation iris
@attribute sepallenght numeric
@attribute sepalwidth numeric
@attribute petallenght numeric
@attribute petalwidth numeric
@attribute class{iris-setosa,iris-veriscolor}
@data
5.1,3.5,1.4,0.2,iris-setosa
4.9,3.0,1.4,0.2,iris-setosa
4.3,3.0,1.1,0.1,iris-setosa
7.0,3.2,4.7,1.4,iris-veriscolor
6.4,3.2,4.5,1.5,iris-veriscolor
6.9,3.1,4.9,1.5,iris-veriscolor
6.3,3.3,6.0,1.9,iris-veriscolor
5.8,2.7,5.1,2.1,,iris-veriscolor
7.1,3.0,5.9,1.8,,iris-veriscolor
Step2: From the weka explorer load the data file iris .arff into weka
Step 3: in order to perform clustering select the cluster tab and click on the choose button. This steps results in dropdown list of available clustering algorithms.
Step 4: In this case we select simplekmeans
Step 5: Now click on the text button to right of the choose button. now a popup window is appeared. In this window select number of clusters.
step 6: now we keep on use training set data and run the algorithm.
Step 7: now we click on the start button to generate the model now the evaluation statistics will appear in the right panel.
Output:
=== Run information ===
Comments
Post a Comment