A Study of Hierarchical Clustering

Clustering has various techniques which can analyze data in efficient manner and generate desired output. Among all clustering technique hierarchical clustering technique is efficient technique which can cluster data in hierarchical manner. In hierarchical clustering, all data is clustered in one cluster. Hierarchical method helps us to cluster the data objects in the form of a tree known as hierarchy. Hierarchical clustering can be performed in two ways: agglomerative clustering and divisive clustering. Agglomerative clustering is always more preferable. For a good cluster analysis, the quality of the clusters should be high. The main problems with hierarchical clustering are accuracy as complex datasets have number of attributes. Due to complex nature of dataset it is very difficult to derive accurate relationship between attributes. When the relationship between attributes is not accurate it leads to reduction in clustering accuracy. In this paper, improvement will be proposed in hierarchical clustering to drive accurate relationship between attributes and improve accuracy of clustering