Matlab toolbox Lesson 11 Matlab toolbox What is
Matlab toolbox Lesson 11
Matlab toolbox • What is a Matlab toolbox? – Toolboxes are specialized collections of Matlab functions built for solving particular classes of problems. • What are they good for? (everything)
A partial list of Matlab’s toolboxes – – – Bioinformatics Communications Financial Image Processing Neural Network Statistics Partial differential equations Symbolic Math Genetic Algorithm GUI Optimization And more…
Clustering (statistics toolbox) • The Matlab statistics toolbox has many features: – Statistical visualization – Probability distributions – Hypothesis testing – Regression analysis – Cluster analysis – And More…
What is clustering? • Clustering is the assignment of a set of observations into subsets (called clusters) so that observations in the same cluster are similar in some sense (Wikipedia) • It is a form of unsupervised learning – you generally do NOT have examples demonstrating how the data should be grouped together
K-means algorithm Clustering Partitional K-means Hierarchical
K-means algorithm • The k-means algorithm partitions the data into k mutually exclusive clusters
K-means algorithm • The k-means algorithm partitions the data into k mutually exclusive clusters
K-means algorithm • The k-means algorithm partitions the data into k mutually exclusive clusters
K-mean algorithm • Randomly choose initial positions of K centroids • Assign each of the points to the “nearest centroid” (depends on distance measure) • Re-compute centroid positions • Stop on converges (loop otherwise)
K-means in matlab • Example on iris data • In the 1920's, botanists collected measurements on the 1) sepal length 2) sepal width 3) petal length 4) petal width • of 150 iris, 50 from each of three species (setosa, versicolor, virginica) • The measurements became known as Fisher's iris data
- Slides: 17