AHC

Caption: Figure 3: Principal component analysis of overall liking ratings of 104 consumers and the average overall liking ratings of consumer clusters from the agglomerative hierarchical clustering analysis.

As a next step, agglomerative hierarchical clustering was used to examine the similarities/dissimilarities between the PGC lines.

[27] developed an improved agglomerative hierarchical clustering algorithm by utilizing Euclidean and Canberra distance measures.

The resulting parameters were analyzed by means of agglomerative hierarchical clustering to explore the feasibility of such an approach.

Agglomerative hierarchical clustering algorithms are exploratory in nature in which initially each item/object is considered as its own cluster and in next steps by repeated application of merging process based on similarity measures the number of clusters finally reduces to single cluster containing all objects.

To conduct agglomerative hierarchical clustering for [FC.sub.intra], the number of initial clusters, n, is determined by analyzing the temporal characteristic between frames.

It provides an agglomerative hierarchical clustering method with computing resemblance coefficients which have quantitative (e.g., location) or qualitative (e.g., connectivity) information types.

Agglomerative Hierarchical Clustering (AHC) was done by using Pearson Correlation Coefficient and Unweighted Pair Group Method with Arithmetic Mean (UPGMA) as Agglomeration method by XLSTAT 2012 version 1.02.

"An unsupervised method of classifying remotely sensed images using Kohonen self-organizing maps and agglomerative hierarchical clustering methods".

ROCK (Guha, 1999), a Robust hierarchical-clustering algorithm is an agglomerative hierarchical clustering based on the notion of links.

Different clustering algorithms are used for clustering the documents such as partitioned clustering (K-means Clustering) and Hierarchical Clustering (Agglomerative Hierarchical Clustering (AHC)).