Improving Hierarchical Clustering with HC-SYM Method
Author Information
Author(s): Chae Minho, Chen James J.
Primary Institution: Division of Personalized Nutrition and Medicine, National Center for Toxicological Research, U.S. Food and Drug Administration
Hypothesis
Can the HC-SYM method improve the ordering of hierarchical clustering to better represent relationships in high-dimensional data?
Conclusion
The HC-SYM method enhances the clarity of relationships between objects in hierarchical clustering and performs better than existing methods.
Supporting Evidence
- The HC-SYM method was shown to improve the clarity of dendrograms in hierarchical clustering.
- Performance metrics indicated that HC-SYM outperformed traditional methods in both local and global perspectives.
- The method can be flexibly applied at various levels of a tree for exploratory analysis.
Takeaway
This study created a new way to organize data that helps us see patterns better, especially when dealing with lots of information.
Methodology
The study proposed the HC-SYM method to order leaves of a binary tree based on bilateral symmetric distances, evaluated through both supervised and unsupervised approaches.
Limitations
The method's performance may vary with different datasets and may not be universally applicable to all types of data.
Participant Demographics
The study analyzed gene expression data from 504 samples across eight tissue types.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website