Hierarchy Clustering

Just like this KMeans clustering, our intention is to create clusters within our dataset, grouping related data so that we may determine different classes or groupings, to allow us to make predictions based on this information in a wide array of applications. However, in the areas in which KMeans fails, Hierarchy Clustering attempts to alleviate the burden somewhat with its several choices of novel techniques such as single-link clustering, or Ward Clustering, Hierarchy Clustering techniques chosen by the user, depending on the layout of their dataset. Hierarchy Clustering at the end of the day is just a regular clustering algorithm, with its advantages and disadvantages, and is by no means the successor of KMeans. Generally, Hierarchy Clustering works as follows; You start off with your dataset, which may be spaced out strange or have a weird layout with strange densities, where clusters are not easily differentiable by you, by all means, you honestly have no idea. That's alright, that usually is the case in real-world problems.