An Extended Agglomerative Hierarchical Clustering Techniques

Abstract

Clustering is a significant method of data analytics in real world environments since human labelling of the data is often costly. Clustering was developed as an alternative to manual tagging. In the field of data analytics, hierarchical clustering is of critical significance, particularly in light of the exponential rise of data derived from the normal world. You may derive a variety of hierarchical agglomerative clustering algorithms from this architecture by providing an inter-cluster semantic similarity, an expression patterns of the -similarity graph, and a cover procedure. These three pieces of information are required. According to the findings of our experiments, our approaches are not only more efficient than conventional hierarchical algorithms, but they also produce smaller agglomerative hierarchical clustering while maintaining the same level of clustering effectiveness. It is generally agreed that topology management is an effective strategy for addressing these challenges. This method groups nodes together for the purpose of managing them and/or carrying out a variety of duties in a dispersed way, such as resource management. There are many quality-driven goals that may be accomplished by clustering, despite the fact that approaches for clustering are mostly renowned for their ability to reduce energy usage. The purpose of this study is to provide a comprehensive explanation on various enhanced agglomerative hierarchical clustering techniques. In addition to this, the authors have provided certain criteria, on the basis of which one may also assess which of these previously described algorithms is the most effective.