Cluster analysis is used to define an item without having to assign a class name to it. It examines all of the data in the Data Warehouse and compares the cluster to the one that is already operational. Its job is to sort a collection of items into groups, often known as clusters. It is used to carry out data mining tasks utilizing statistical data analysis techniques. It contains all of the knowledge and information on a variety of areas, including Machine Learning, Pattern Recognition, Image Analysis, and Bioinformatics. Cluster analysis involves tries and failures in the iterative process of knowledge development. It is used in conjunction with pre-processing and other factors to obtain the desired characteristics.
Purpose of cluster analysis:
- Ability to deal with different kinds of attributes
- Discovery of clusters with attribute shape
- High dimensionality
- Ability to deal with noise