So far, we have primarily explored supervised learning methods where we have a historical trail of data that is used for training the machine learning models. However, there is a very common scenario where the machine needs to classify objects or entities into various groups based on predefined or runtime categories. For example, in the dataset that contains information about employees, we need to categorize the employees based on one or more attributes combined. With this, the goal is to group similar objects and partition the data based on similarities.
The general idea is to have a consistent attribute map within a group and distinct behaviors across the groups. Unlike the supervised learning methods, there are no dependent variables in the case of data clustering. A cluster represents various groups of entities that demonstrate similarities in attributes...
