Introduction to WEKA installation and a simple example

1. Download from http://www.cs.waikato.ac.nz/ml/weka for Windows, Mac, Linux
2. Installation, changeable path

   Anhao, as shown below

3. Run

   Open erxplore.

   Click Open file to open the data file.

   Click a data file in the installation directory, such as the data folder, to copy it to a common folder for easy use.

   Open a data file in the data folder, for example, weather. Nominal

   Then, look at the various attributes and tag buttons on this interface and become familiar with them.

   Click the edit button. The following table is displayed, which is very intuitive.

   The content of this data file indicates whether to play in different weather conditions.

   A total of 14 instances (days), five attributes (weather appearance, temperature, humidity, no wind, whether to play), the first four are features, the last is classification results. The numeric fields in this example are discrete.

The following sets of data are used to train the decision tree.

Click classify, click choose, select j48 from the tree, select the cross-validation method, and click Start. The result is displayed on the right.

Right-click the selected experiment and select visualize tree.

This is the trained decision tree, and the recognition rate is 50% from the right display box. The results of this example are largely due to the limited number of training samples.

 

Well, WEKA's installation and use of simple examples are like this. WEKA is still very powerful. Learn more.

 

 

Introduction to WEKA installation and a simple example