PROJECT TITLE :

Hierarchical Clustering Given Confidence Intervals of Metric Distances - 2018

ABSTRACT:

This Project considers metric the exact dissimilarities between pairs of points aren't unknown but known to belong to some interval. The goal is to review methods for the determination of hierarchical clusters, i.e., a family of nested partitions indexed by a resolution parameter, induced from the given distance intervals of the dissimilarities. Our construction of hierarchical clustering strategies is predicated on defining admissible ways to be those strategies that satisfy the axioms of value-nodes in an exceedingly metric area with two nodes are clustered together at the convex combination of the higher and lower bounds determined by a parameter-and transformation-when both distance bounds are reduced, the output might become a lot of clustered however not less. 2 admissible ways are made and are shown to produce universal bounds in the house of admissible strategies. Practical implications are explored by clustering moving points via snapshots and by clustering coauthorship networks representing collaboration between researchers from different communities. The proposed clustering ways succeed in identifying underlying hierarchical clustering structures via the maximum and minimum distances in all snapshots, with in differentiating collaboration patterns in journal publications between different analysis communities based on bounds of network distances.