clustering2_4perPage

clustering2_4perPage - Outline Clustering 1 Announcements...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Clustering Data Mining Prof. Dawn Woodard School of ORIE Cornell University 1 Outline 1 Announcements 2 Clustering 2 Announcements Questions? 4 K-Means Clustering Last time we discussed initializing k-means by choosing random cluster assignments for each of the points In practice k-means often converges more quickly if we instead start by choosing the cluster centers { m k : k = 1 ,..., K } Those cluster centers imply a cluster assignment C so this is equivalent 6
Background image of page 1

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

View Full DocumentRight Arrow Icon
K-Means Clustering Figure 14.6 in our text: 7 K-Means Clustering 8 K-Means Clustering 9 Public Utilities Data Recall the data on public utilities (Shmueli, Patel, and Bruce, 2007) Wish to group based on fnancial ±actors Used ±or e.g. a study on the impact o± deregulation Could pick one “typical” utility in each group and study in detail the potential e±±ect o± deregulation on that utility Scale up to estimate impact ±or all utilities This is less costly than studying in detail the e±±ect o± deregulation on every single utility 10
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 12/23/2009 for the course ORIE 4740 at Cornell.

Page1 / 7

clustering2_4perPage - Outline Clustering 1 Announcements...

This preview shows document pages 1 - 3. Sign up to view the full document.

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