comparing Gaussian mixtures and k-means
Summary
Gaussian mixture models and K-means are two canonical approaches to clustering, i.e. dividing data points into meaningful groups. This concept node discusses the tradeoffs between them.
Context
This concept has the prerequisites:
- mixture of Gaussians models
- k-means
- Expectation-Maximization algorithm (K-means is analogous to the EM algorithm for fitting Gaussian mixtures.)
Core resources (we're sorry, we haven't finished tracking down resources for this concept yet)
Supplemental resources (the following are optional, but you may find them useful)
-Free-
→ Bayesian Reasoning and Machine Learning
-Paid-
→ Pattern Recognition and Machine Learning
A textbook for a graduate machine learning course, with a focus on Bayesian methods.
Location:
Section 9.3.2, pages 443-444
→ Machine Learning: a Probabilistic Perspective
A very comprehensive graudate-level machine learning textbook.
Location:
Section 11.4.2.5-11.4.2.7, pages 352-355
See also
- Other clustering methods include:
- naive Bayes , a probabilistic model for discrete data
- spectral clustering , for when the clusters don't have nice convex shapes