Clustering and Classification

Course Description

Clustering and Classification methods are used to determine the similarity or dissimilarity among samples. Clustering methods are usually exploratory analysis methods used in an unsupervised way which elucidate the similarity within a set of samples. They are often used to determine if there are natural groupings and/or particularly unique individuals or groups within a set. Classification, on the other hand, is a supervised method which uses data with known group assignments and attempts to determine which group(s), if any, a new sample belongs to. This course will discuss various clustering and classification methods and the practical considerations for using them. The course includes hands-on computer time for participants to work example problems using PLS_Toolbox or Solo.

Prerequisites

Chemometrics I–PCA and Chemometrics II–Regression and PLS or equivalent experience.

Course Outline

1. Linear Discriminant Analysis
   
2. K-Nearest Neighbors (KNN)
   
3. K-Means
   
4. SIMCA
   
5. PLS Discriminant Analysis
   
6. Complex Classification Problems