Clustering means grouping items that are similar to each other. Grouping similar products, grouping similar articles or documents, and grouping similar customers for market segmentation are all examples of clustering. The core principle of clustering is minimizing the intra-cluster distance and maximizing the intercluster distance. The intra-cluster distance is the distance between data items within a group, and the inter-cluster distance is the distance between different groups. The data points are not labeled, so clustering is a kind of unsupervised problem. There are various methods for clustering and each method uses a different way to group the data points. The following diagram shows how data observations are grouped together using clustering:
As we are combining similar data points, the question that arises here is how to find the similarity between two data points so we can group similar data objects into the same cluster. In order to measure the similarity or dissimilarity...