Logistic Regression: not dead yet
Talk Information
Mountain View, CA, 7/28/05
Tags
Abstract
These slides were presented at Google, Inc, on 28 July 2005. The first half of these slides discusses why we like LR for data mining, and how we accelerate parameter fitting. The second half discusses several interesting and unusual applications of LR, most of which have software or papers available from the Auton lab website and the author's website.
Attachment Data
Download (application/pdf, 2.3 MB)
Links
- Paper: Making Logistic Regression A Core Data Mining Tool With TR-IRLS
- Paper: Logistic Regression for Data Mining and High-Dimensional Classification
- Paper: Making Logistic Regression A Core Data Mining Tool: A Practical Investigation of Accuracy, Speed, and Simplicity
- Software: LR using TR-IRLS, source and binaries available under the GNU GPL
- Software: Sparse Logistic Regression
- Software: Dense Logistic Regression