Unsupervised Learning may be a class of Machine Learning techniques to seek out the patterns in data. the information given to the unsupervised algorithm doesn’t seem to be labeled, which means only the input variables(x) are given with no corresponding output variables. In unsupervised learning, the algorithms are left to themselves to get interesting structures within the data.
The 2 main varieties of Unsupervised learning is
Cluster Analysis: Clustering is that the task of dividing the population or data points into a variety of groups specified data points within the same groups are more the same as other data points within the same group as those in other groups.
Dimensionality reduction technique: One amongst the popular dimensionality reduction techniques is a principal component analysis that is employed in the correlation analysis.
Factor Analysis may be a technique applied for data reduction. The target of correlation analysis is to cut back and summarize an outsized number of an original variable into a smaller number of variables called factors. the fundamental idea of correlation analysis as an information reduction technique is to mix two or more correlated variables into one Factor.
Examples for Unsupervised learning:
§ Hierarchical Clustering
§ K-means Clustering
Hierarchical Clustering: Hierarchical Clustering is an algorithm that builds a hierarchy of clusters. This algorithm starts with all the information points assigns to a cluster of their own. Then two nearest clusters are merged into an identical cluster. In the end, this algorithm terminates when there’s only a single cluster left.
The results of hierarchical closeting are often shown by using Dendrogram.
Steps for hierarchical clustering:
1. Start with N clusters, one for every datum (each data point)
2. Merge the 2 clusters that are closest to every other. Now you’ve got N-1 clusters.
3. Re-compute the distances between the clusters. There are several ways to try and do this. One of them is to think about the gap between two clusters to be the common distance between all their respective members
4. Repeat steps2 and steps3 until you get one cluster of N data points. You get a tree.
K-MEANS CLUSTER: The most popular non-hierarchical approach is the K-mean cluster method.
This process doesn’t involve a tree-like construction process instead it assigns the thing into clusters.
In K-means clustering the way these groups are defined is by creating a centroid for each group. The centroids are just like the heart of the cluster they capture the points closest to them and add them to the cluster.
Steps to K-means clustering:
1. Define the k centroids.
2. Find the closest centroid and update the cluster assignments
3. Move the centroids to the middle of their clusters
What are different types of Unsupervised learning in machine learning?
What is unsupervised learning in ML?
How does unsupervised machine learning work?