[editorskit display=”wordcount” before=”Reading Time: ” after=” min”]
In Unsupervised Learning, the variables and data patterns are not classified. Instead, the machine must discover the hidden patterns and create labels through the use of unsupervised learning algorithms. The k-means clustering algorithm is a popular, yet simple algorithm which an example of unsupervised learning that groups data points that are found to possess similar features. Thus, in unsupervised learning we have only input data (x) but no corresponding output variables (y).
These are called unsupervised learning because unlike supervised learning there is no correct answers (output variables) or there is no teacher. Unsupervised learning algorithms, on their own cognition devise and create patterns and differences without any prior training of data. That means no training will be given to the machine.
The advantage of unsupervised learning is that it enables us to discover patterns in the data that really existed which we were unaware of. The unsupervised learning is particularly powerful during fraud detection as those attacks are not yet labelled or classified. The k-means clustering technique helps in further analysis after discrete groups have been discovered. Unsupervised learning related problems can be further classified and grouped into clustering and association problems.
A clustering problem is where you want to discover the inherent groupings in the data such as grouping customers by purchasing behavior.
Association: An association rule learning problem is where you want to discover rules that describe large portions of your data such as people that buy X and also tend to buy Y.