✦ScienexySCIENCE KNOWLEDGE LIBRARY
Science reference · SCI-CS-003-084

Hierarchical Clustering

Hierarchical clustering organizes observations into a hierarchy of nested clusters that can be visualized using a tree-like diagram called a dendrogram.

Explanation

Hierarchical clustering is an unsupervised learning method that builds a hierarchy of clusters. Agglomerative clustering follows a bottom-up approach: each observation begins in its own cluster, and the closest clusters are repeatedly merged. Divisive clustering follows a top-down approach by starting with one large cluster and repeatedly dividing it into smaller groups. Cluster similarity is determined using a linkage criterion, such as single, complete, average, or Ward linkage. A dendrogram represents the merging or splitting structure and the associated distances. A user can select a desired number of clusters by cutting the dendrogram at an appropriate level. Hierarchical clustering does not always require the number of clusters to be specified at the beginning, but it can be computationally expensive for large datasets and may be sensitive to distance and linkage choices.

Example or application

A researcher groups different plant species based on measured characteristics. A dendrogram shows which species are most similar and how smaller groups combine into larger groups.

Text

Hierarchical clustering creates a multilevel structure showing how individual observations or groups combine or divide.

Key Points

  • Hierarchical clustering creates nested groups.

  • Agglomerative clustering uses a bottom-up approach.

  • Divisive clustering uses a top-down approach.

  • A dendrogram visualizes the clustering hierarchy.

  • Linkage criteria determine how cluster distances are calculated.

  • The final number of clusters can be selected by cutting the dendrogram.

Applications

  • Biological taxonomy

  • Gene expression analysis

  • Document organization

  • Customer segmentation

  • Exploratory data analysis

Learning Outcomes

  • Define hierarchical clustering.

  • Differentiate agglomerative and divisive approaches.

  • Explain the purpose of a dendrogram.

  • Compare common linkage criteria.

Practice Questions

  • What is hierarchical clustering?

  • How does agglomerative clustering work?

  • What is a dendrogram?

  • How do single and complete linkage differ?

Related Concepts

  • Clustering

  • Agglomerative Clustering

  • Divisive Clustering

  • Dendrogram

  • Linkage Criteria

Common Mistakes

  • Confusing agglomerative and divisive clustering.

  • Assuming every linkage method produces the same hierarchy.

  • Interpreting dendrogram height without understanding the distance measure.

  • Ignoring computational cost on large datasets.

Sources and verification

Sources are shown as provided in the project data. Not every source has yet been independently verified. Please report errors or missing citations.

  • Murtagh, F., and Contreras, P. (2012). Algorithms for Hierarchical Clustering: An Overview.
  • Hastie, T., Tibshirani, R., and Friedman, J. (2009). The Elements of Statistical Learning.

Suggest a correction or source · More in Artificial Intelligence & Machine Learning