|
|
Guests are welcome to view our materials. To subscribe, edit, view raw markup, etc., you'll need to register for an account. Accounts are free (and will always be free) - your involvement helps us directly and indirectly (by demonstrating that our work matters to our funders...) StartingPoints has more info.
PatternRecognitionSpr09ClassRoster
UsingSafariBooks
What's New
- Having problems getting going with e1071? Take a look at the tutorial listed in the Support Vector Machine resources below!
- AgileR
- New material on packaging (critical to successfully developing R code in a reasonable amount of time with a reasonable amount of effort
 on the RisCool topic.
Orientation
Course Description
Syllabus
Rather than trying to do an exhaustive review of statistical pattern recognition techniques, we'll focus on one major unsupervised technique, self-organizing maps, and one major supervised technique, support vector machines, to give you a chance to use each in some depth.
- Samuel Kaski's doctoral thesis on Data Exploration Using Self-Organizing Maps, see especially Chapters 2 (Methods for exploratory data analysis) and 3 (Stages of SOM-based exploratory data analysis).
- SOM in the wild, or an example of how SOM are actually used in a real project: Unobtrusive User Modeling for Adaptive Hypermedia (preprint)
- Mat Buckland's SOM tutorial
- SOM Bibliography from Helsinki University of Technology
- multiple contributed SOM packages in R, as well as multiple packages that include SOM
- package list:
-
kohonen - a fairly modern implementation, as well as fairly complete (although missing some crucial elements, like sammon maps.) Includes a nice adaptation to using SOMs for classification.
-
wccsom - builds upon kohonen, adding weighted cross-correlation. Very nice, indeed.
-
som - a bare bones SOM implementation, has quantization error, but no Umatrix, no component maps.
-
class - contains batchSOM implementation (from 1995!).
- reflective question: how do you decide which package to use? Some suggested criteria:
- package quality
- reflective question: how do you assess the quality of a package? Some suggested criteria:
- does it include unit tests?
- how comprehensive is the documentation?
- ... (what else)?
- what is the goal of your overall project? does your goal match the goal of the package?
- ... (what else)?
- Support Vector Machines for Classification and Regression - by Steve Gunn, is a more general discussion of SVMs, focusing on use and how they fit in to the bigger picture.
- A Tutorial on Support Vector Machines for Pattern Recognition - by Christopher J.C. Burges, has lots more detailed info on the different kernels and how you decide which to use. It also has excellent figures illustrating the optimal separating hyperplane for various types of kernels.
- lots of useful info at Support Vector Machines dot org, including a lovely visual explanation of the optimum hyperplane
- R resources for SVMs are listed in the Machine Learning task view at cran
- More on
e1071 (recommended, as it has all sorts of cool stuff)
- Here's a nice tutorial from Hannes Planatscher and Janko Dietzsch
- e1071 listing on crantastic. You can retrieve the documentation, search the r-help archives, etc.
- you'll almost surely need to download and install libSVM separately. I built the C++ executables, rather than the Java ones, on my mac, as follows:
- I downloaded the
.tar.gz archive, and moved it to my user account.
- I opened a terminal window and unpacked the archive from the command line
- I ran make.
- My
R installation found libsvm without any further configuration. Nice.
- There's a
Makefile.win for the windowz types out there...
How Things Work
Deadlines
- SvMLabSpr09
- rough draft due 6/9/2009 midnight
- peer eval due 6/11/2009 midnight
- final version (firm) due 6/13/2009 midnight
- I have to have grades in Tuesday morning, so I won't be able to be very flexible on the deadlines. A few hours doesn't matter, but a few days is impossible!
- SomLabSpr09
- rough draft due 5/6/2009 midnight
- peer eval due 5/9/2009 midnight
- final version (tentative) due 5/14/2009 midnight
- note that you'll have to start working on the support vector machine stuff right away...
- PlayingWithData
- rough draft due 4/20/2009 midnight
- peer eval due 4/22/2009 midnight
- final version due 4/29/2009 midnight
Class Materials
Class Notes
References
Tools
Assignments
Preps
Labs
|
|
Guests are welcome to view our materials. To subscribe, edit, view raw markup, etc., you'll need to register for an account. Accounts are free (and will always be free) - your involvement helps us directly and indirectly (by demonstrating that our work matters to our funders...) StartingPoints has more info.
|
Copyright 1999-2009 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.