Data Mining with WEKA

Exercise Instructions

1. Convert Your Dataset to ARFF Format

If you already have a dataset in CSV, Excel, or another format, you can convert it to ARFF using Weka’s built-in tools or online converters:

Using Weka GUI:

• Open Weka: Launch the Weka GUI.
• Preprocess: Click on the "Explorer" button.
• Open File: In the Preprocess tab, click on "Open file..." and select your dataset (e.g., CSV file).
• Save as ARFF: After loading the data, go to "File" > "Save as..." and choose ARFF format to save your dataset.

Using Weka Command Line:

To convert a CSV file to ARFF via the command line, use the following command:

java -cp weka.jar weka.core.converters.CSVLoader yourfile.csv > yourfile.arff

2. Creating ARFF Files Manually

If you want to create an ARFF file manually, you can use a text editor. Here’s the general structure of an ARFF file:

@RELATION dataset_name
@ATTRIBUTE attribute1_name ATTRIBUTE_TYPE
@ATTRIBUTE attribute2_name ATTRIBUTE_TYPE
...
@ATTRIBUTE class_attribute_name {class1, class2, ...}

@DATA
value1, value2, ..., class_value
value1, value2, ..., class_value
...
      

3. Example of ARFF File Structure

Here’s an example of a simple ARFF file:

@RELATION iris

@ATTRIBUTE sepal_length NUMERIC
@ATTRIBUTE sepal_width NUMERIC
@ATTRIBUTE petal_length NUMERIC
@ATTRIBUTE petal_width NUMERIC
@ATTRIBUTE class {Iris-setosa, Iris-versicolor, Iris-virginica}

@DATA
5.1, 3.5, 1.4, 0.2, Iris-setosa
4.9, 3.0, 1.4, 0.2, Iris-setosa
4.7, 3.2, 1.3, 0.2, Iris-setosa
...
      

4. Using ARFF in Weka

Once you have your ARFF file ready:

• Open Weka and go to the Explorer.
• Click on "Open file..." and select your ARFF file.
• You can now explore data and apply different algorithms using Weka's tools.

For more information, refer to Weka Documentation.