Clustering is a methodology that you can use in order to search for consistent groups of data within a given dataset. Using comparison techniques, we can look for groups of items within the dataset that contain similar characteristics. These individual datapoints are then divided into clusters. Clustering is commonly used in order to solve multi-objective problems.
There are two general classifications of clustering, both of which have distinct subclassifications:
- Hard clustering: The datapoints within the dataset are either explicitly a part of a cluster or not explicitly part of a cluster. Hard clustering can be further classified as follows:
- Strict partitioning: An object can belong to exactly one cluster.
- Strict partitioning with outliers: Strict partitioning, which also includes a concept that objects can be classified as outliers (meaning they belong...