cluster1_article

cluster1_article - 1 Classification and clustering how they...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 1 Classification and clustering: how they differ Classification and clustering • Classification and clustering are superficially similar • It is easy to confuse them • Yet ultimately they are rather different • Both assume a population is divided into subpopulations – called “classes” or “clusters” Classification • In classification: – the classes are defined a priori – there is a training sample where the classification is known – for this reason called supervised learning • Examples: – we have a sample of castings where it is known whether each casting is usable – there is a sample where we know whether each customer made a purchase – for a sample of dealers we know whether each dealer pur- chases in batches Classification by logistic regression • Logistic regression is a popular and powerful classification tool • We have studied logistic regression when there are only two classes – Multivariate logistic regression allows three or more classes * Covered in more advanced courses Clustering • Goal: to divide the population into homogeneous subpopula- tions (clusters) • The locations of the clusters are unknown • The number of the clusters might also be unknown – The number of clusters will depend on how much splitting one wishes to do • No training sample is available – Clustering is an example of unsupervised learning • there is no single “ correct answer ” 2 Examples of clustering Examples of clustering • Customer segmentation: – Divide population of customers into segments with similar needs, attitudes, and behaviors – Then tailor marketing to specific segments 2 Examples of clustering • Cluster human tumors by gene expression: – in one example there were 64 tumors and 6830 genes – clusters were compared with prior classification of tumors – clustering helped detect two misdiagnoses Examples of clustering • Vector quantization (for image compression) – 2 × 2 blocks of pixels are clustered – each block is replaced by its cluster average – only cluster averages and cluster memberships are saved – the number of clusters determines the amount of compres- sion 3 Simulated data example Example: simulated data ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●-2 2 4 6-2 2 4 6 x 1 x 2 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●...
View Full Document

This note was uploaded on 01/09/2009 for the course ORIE 312 taught by Professor D.ruppert,p.jacks during the Spring '08 term at Cornell.

Page1 / 13

cluster1_article - 1 Classification and clustering how they...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online