 VECA is an open source data mining tool, it is developed by Waiketo University, New Zealand. It can be downloaded from the following link and it runs on Windows, Mac operating system and Linux. This is the VECA GUI chooser. It has five applications, Explorer, Experimenter, Knowledge Flow, Workbench and Simple CLI. For most of our work, we will choose to work on Explorer. The Explorer has the following menus, Pre-Process, Classify, Cluster and Visualize. We can open the file either saved on your system or we can also use URL. For demonstration purposes, we will choose the file that you get when you download the VECA software. And this file can be located in C drive program files and VECA. This is the sample data file. For demonstration, we will use the Iris data set. This is the specifications of the given Iris data set. It has five attributes that is columns and 150 instances that is rows. It has three classes, Iris Satoshi, Iris Versicolor, Iris Virginica is having 50 counts. It also shows the petal width, petal length, sample width and sample length variation across the different attributes or classes. We can also visualize the given data set by choosing the visualize menus. Here we can visualize our given data set. In this example, the sample length is plotted along x-axis and the class level is on y-axis. This plot shows the variation of class with respect to sample length or we can also choose the different parameter along the x-axis suppose we choose petal width and along the y-axis we can choose petal length. This plot shows the variations of different classes along petal width and petal length. Now let's learn about classification in the VECA tool. For classification also, we choose the explorer application and we choose the downloaded data sets from our system which is stored in the CFI. In VECA, we have different classifier. First we choose the classified menu and choose the appropriate classifier. In the VECA, we have the different classifier. We have the Navier-based classifier. We have the IBK classifier. We have the zero classifier. Zero classifier is also known as the trivial classifier or baseline classifier. We also have the J48 classifier. For our analysis purpose, we will choose the J48 classifier. Before analyzing the data, we create a model. The first we divide the data into two sets, training set and testing set. The training set is used to create a classifier model and the test data set can be used to check the performance of the model created. For most of the application, two-third of the given data set is selected for training the model and the remaining one-third data set is used for testing the model. And now we run. This is the classifier output and we can obtain the summary as following. We are having 51 as the total number of instances. This belongs to the testing set out of which 49 instances are classified. As classified as correct and only two instances as classified as incorrect. We can also see the KEPA statistic, mean absolute error values, root mean square error value, relative absolute error values and etc. Here we can see the detailed accuracy of the classes. We have the value of the precision which equals to 0.965, we have the value of recall, we have the average value of the F-measures, F1, 0.961, we also have the values of ROC area. And on the left-hand side at the bottom you can see the confusion matrix. Here we can see that for the iris ratosa all the instances are classified as correct. No one instances as classified as incorrect one. Same is the case for the iris versicolor here. Here also all the instances are classified as correct and no other instances is classified as incorrect. But for the iris virginica out of 17 instances, 15 instances are classified as correct and two instances are the classified as incorrect.