Deep dive into unsupervised learning algorithms

Unsupervised machine learning deals with learning unlabeled data—that is, data that has not been classified or categorized, and arriving at conclusions/patterns in relation to them.

These categories learn from test data that has not been labeled, classified, or categorized. Instead of responding to feedback, unsupervised learning identifies commonalities in the data and reacts based on the presence or absence of such commonalities in each new piece of data.

The input given to the learning algorithm is unlabeled and, hence, there is no straightforward way to evaluate the accuracy of the structure that is produced as output by the algorithm. This is one feature that distinguishes unsupervised learning from supervised learning. 

What does it mean to learn without an objective? Consider the following:

• Explore the data for natural groupings.
• Learn association rules, and later examine whether they can be of any use.

Here are some classic examples:

• Performing market basket analysis and then optimizing shelf allocation and placement
• Cascaded or correlated mechanical faults
• Demographic grouping beyond known classes
• Planning product bundling offers

In this section, we will go through the following unsupervised learning algorithms with easy-to-understand examples:

• Clustering algorithms
• Association rule mapping