An introduction to classification algorithms
Classification is the process of categorizing any kind of entity or class so that it is better understood and analyzed. The classifying process usually happens as part of a pre-setup business process (for example, tagging a product as defective or good after observing it), or through a return process (for example, tagging a product as defective after the customer returned it as defective). In either event, the important point is classifying an entity – in this case, a product into a class (i.e., defective or not).
Figure 6.1 shows data that has been classified into two classes using three input variables. The figure shows where a pair of Input and Output data points are categorized into two classes. When output labels consist of only two classes, it is called a binary classification problem:
Figure 6.1 – Binary classification
If the output variable consists of more than two classes –...
