Data Mining Tutorial

                 

1.Weka Quick Start :

   

•     Download & Install Weka :    

>    Download Weka for linux , for Windows . An How to open it?

•     WEKA Explorer Environement    :

>    Explorer Environement under weka

2.  Weka labs:

   

•      LAB1:Classification(Supervised):

>    with weka.

•     LAB2:Decision Tree (Example of Supervised):(Not yet)

>    with weka . 

•     LAB3:Simple K-Means(Unsupervised):  (not yet)  

>   with weka

Lab3: Simple K-Means(Unsupervised): (With Weka)

TO DO (Not Yet)

Lab2 : Decision Tree

TODO . (NOT YET)
classify >J48>Cross Validation

=== Run information ===

Scheme:weka.classifiers.trees.J48 -C 0.25 -M 2

Relation:     iris

Instances:    150

Attributes:   5

              sepallength

              sepalwidth

              petallength

              petalwidth

              class

Test mode:10-fold cross-validation

=== Classifier model (full training set) ===

J48 pruned tree

------------------

petalwidth <= 0.6: Iris-setosa (50.0)

petalwidth > 0.6

|   petalwidth <= 1.7

|   |   petallength <= 4.9: Iris-versicolor (48.0/1.0)

|   |   petallength > 4.9

|   |   |   petalwidth <= 1.5: Iris-virginica (3.0)

|   |   |   petalwidth > 1.5: Iris-versicolor (3.0/1.0)

|   petalwidth > 1.7: Iris-virginica (46.0/1.0)

Number of Leaves  :  5

Size of the tree :  9

Time taken to build model: 0.04 seconds

=== Stratified cross-validation ===

=== Summary ===

Correctly Classified Instances         144               96      %

Incorrectly Classified Instances         6                4      %

Kappa statistic                          0.94

Mean absolute error                      0.035

Root mean squared error                  0.1586

Relative absolute error                  7.8705 %

Root relative squared error             33.6353 %

Total Number of Instances              150    

=== Detailed Accuracy By Class ===

               TP Rate   FP Rate   Precision   Recall  F-Measure   ROC Area  Class

                 0.98      0          1         0.98      0.99       0.99     Iris-setosa

                 0.94      0.03       0.94      0.94      0.94       0.952    Iris-versicolor

                 0.96      0.03       0.941     0.96      0.95       0.961    Iris-virginica

Weighted Avg.    0.96      0.02       0.96      0.96      0.96       0.968

=== Confusion Matrix ===

  a  b  c   <-- classified as

 49  1  0 |  a = Iris-setosa

  0 47  3 |  b = Iris-versicolor

  0  2 48 |  c = Iris-virginica

-if confidence Factor discreased , We have

=== Confusion Matrix ===

  a  b  c   <-- classified as

 49  1  0 |  a = Iris-setosa

  0 46  4 |  b = Iris-versicolor

  0  4 46 |  c = Iris-virginica

==>if confidence Factor discreases  Error numbers increases

20.png

  a  b  c   <-- classified as

 15  0  0 |  a = Iris-setosa

  0 19  0 |  b = Iris-versicolor

  0  2 15 |  c = Iris-virginica

=========

kmeans

1)with 2 clusters:

=== Run information ===

Scheme:weka.clusterers.SimpleKMeans -N 2 -A "weka.core.EuclideanDistance -R first-last" -I 500 -S 10

Relation:     iris

Instances:    150

Attributes:   5

              sepallength

              sepalwidth

              petallength

              petalwidth

              class

Test mode:evaluate on training data

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

kMeans

======

Number of iterations: 7

Within cluster sum of squared errors: 62.1436882815797

Missing values globally replaced with mean/mode

Cluster centroids:

                                          Cluster#

Attribute                Full Data               0               1

                             (150)           (100)            (50)

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

sepallength                 5.8433           6.262           5.006

sepalwidth                   3.054           2.872           3.418

petallength                 3.7587           4.906           1.464

petalwidth                  1.1987           1.676           0.244

class                  Iris-setosa Iris-versicolor     Iris-setosa

Time taken to build model (full training data) : 0.03 seconds

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

Clustered Instances

0      100 ( 67%)

1       50 ( 33%)

2)With 3 clusters:

=== Run information ===

Scheme:weka.clusterers.SimpleKMeans -N 3 -A "weka.core.EuclideanDistance -R first-last" -I 500 -S 10

Relation:     iris

Instances:    150

Attributes:   5

              sepallength

              sepalwidth

              petallength

              petalwidth

              class

Test mode:evaluate on training data

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

kMeans

======

Number of iterations: 3

Within cluster sum of squared errors: 7.817456892309574

Missing values globally replaced with mean/mode

Cluster centroids:

                                          Cluster#

Attribute                Full Data               0               1               2

                             (150)            (50)            (50)            (50)

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

sepallength                 5.8433           5.936           5.006           6.588

sepalwidth                   3.054            2.77           3.418           2.974

petallength                 3.7587            4.26           1.464           5.552

petalwidth                  1.1987           1.326           0.244           2.026

class                  Iris-setosa Iris-versicolor     Iris-setosa  Iris-virginica

Time taken to build model (full training data) : 0.01 seconds

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

Clustered Instances

0       50 ( 33%)

1       50 ( 33%)

2       50 ( 33%)

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

Wich Attributs for better Classification in kmeans ?

=> petallength(inter/ intra Class)

28_intraInter_patellength.png

=> petallewidth(inter/ intra Class)

29_intraInter_patellength.png

Lab1: Classification (Supervised): (With Weka)

In this lab you will learn about using weka using the database iris.arff you find it  in the folder: ~ / weka-3-6/data/Iris.arff.

This is a basic donnnées, very famous, with 150 examples of flowers described by four attributes of continuing value and belonging to three classes.

1.DATA :

1>Open firstly file Iris.arff with a text editor to discover the format ARFF (Attribute Relation File Format) :

2>Click on the button "O". Choose the data file. / Data / Iris.arff:

Some information will appear in the window

In the pane "SelectedAttribute", you can get basic statistics for the selected attribute: Name, Type, Missing, Unique, distinct, Vlaur Min / Max.

Select SepalLength Attribut:

Select SepalWidth ATTribut  : 

Select PetalLength :

Select PetalWidth :

Select Class:

So we have : 

     >>>Blue Color => Iris-Setosa

     >>>Red Color => Iris-versicolor

     >>>Last Color => Iris-virginica

-Click on Visualize All : 

==>We Note that the overlap(chevauchement(fr))  between classes is minimal when we have classification based on attribute petal (Either  petalLength or petalWidth).

-Weka offers the possibility to operate by pretreatment by applying a filter on attributes:

- Click on Choose Button >supervised>attribute>Discretize :

==>This filter allows to discretize continuous values

2.Data visualization :

For a first approach to classification, go in the "Visualize" Tab. You will see a set of 25 graphics:

1. Change the axes to achieve a classification that gives off the lower of decision rules

-When X-> PetalWidth && Y -> SepalLength , We give off a minimal number of  Decisions Rules : 

Decisions Rules (Règles de décision(fr)) :

 -if(X>=0.1) AND(X<=0.6) Then Class<-- Iris-Setosa.

-if(X>=1) AND(X<=1.7)  AND(Y<=5.6) Then Class<-- Iris-Versicolor.

-if(X>=1.8) AND(X<=2.5)  Then Class<-- Iris-virginica.

-if(Y>=7.1)  Then Class<-- Iris-virginica.

......

=>This is the best classification in comparison with other classifications. In effect, the overlap between classes is minimal

Tuto 2 :WEKA Environment Explorer

In the previous tuto ,we Have seen how to install WEKA 

1.Open weka.jar & click on Explorer : 

A new window opens "WEKA Knowledge Explorer" with six tabs:

a.Preprocess: to select a file, inspect and prepare the data.

b.Classify: To CHOOSE, implement and test different classification algorithms: in our case this would be a decision tree algorithm.

c.Cluster: To Choose, apply and test the segmentation algorithms.

d.Associate: To apply the algorithm for generating association rules.

e.Select Attributes: To choose the most promising attributes.

f.Visualize: to display (in two dimensions) certain attributes in other functions.

Tuto1 : "Weka" => Quick Start

1.Download Weka :
       a.For linux : Click Here to download
       b.For Windows : Click Here to download.
2.Open Weka:
       a.With linux (Ubuntu): 

•     Open terminal in ./weka-3-6-0/ folder .
•     $  PATH=$PATH:.
•     $ chmod 777 *
•     $ weka.jar &

       
 And you will get this UI : 

       

     b.With Windows :

•                     install weka .
•                     go to <%PATH_OF_INSTALLATION%>\weka-x-y\
•                      Double Click on weka.jar

                  And you will get this UI:

3. Weka Environments:

   After you run it, you get the window title WEKA GUI CHOOSER with the following environments:   

    a.Explorer:  

                                an environment for exploring data    
      b. Experimenter:
                  an environment to perform experiments and statistical tests between the learning patterns.
      c.KnowledgeFlow :it's such as Explorer but it's an environment with "Drag &Drop" UI and
      d.Simple CLI :
                                command line interface.