Answered
FLD (Fisher Linear Discriminant)
If you have MATLAB 11b, take a look at class ClassificationDiscriminant. If you have an older version, take a look at function C...

14 years ago | 0

Answered
classify
"Pooled" and "stratified" are conventional terms in the statistics literature. A pooled estimate is computed using all observati...

14 years ago | 0

| accepted

Answered
FISHER DISCRIMINANT FUNCTION - ROC
Open a documentation window by typing 'doc' in your session or selecting Product Help on MATLAB desktop. Then type 'roc' into th...

14 years ago | 0

Answered
Haar feature based cascade classifiers and duplex flexible window
Sorry, I am not going to look at the paper you mention. FWIW, I suspect that Harr features is what you extract from an image usi...

14 years ago | 0

Answered
Fisher Linear Discriminant
Fisher discriminant is provided in the Statistics Toolbox. If you have MATLAB 11b, use ClassificationDiscriminant. In an earlier...

14 years ago | 0

| accepted

Answered
TreeBagger gives different results depending on 'oobvarimp' being 'on' or 'off'
One way of assessing what is high and low under these circumstances would be to look at the classification error. It can be mode...

14 years ago | 0

Answered
TreeBagger gives different results depending on 'oobvarimp' being 'on' or 'off'
Computing variable importance by permuting observations across every variable (that's what you get when you set oobvarimp to on)...

14 years ago | 0

| accepted

Answered
How is varimportance calculated?
http://www.mathworks.com/help/toolbox/stats/classregtree.varimportance.html If you split parent node 1 into child nodes 2 and...

14 years ago | 0

| accepted

Answered
Decision/Regression stumps using ClassificationTree.template or RegressionTree.template
You get stumps by default for any boosting algorithm from fitensemble. Or, equivalently, you can set 'minparent' to the number o...

14 years ago | 0

| accepted

Answered
Compare two trees generated using same classes but different attributes in each case
There may be a straightforward way, but it is up to you to define criteria for such a comparison. Here is a list of all tree met...

14 years ago | 0

| accepted

Answered
Classification and regression trees
Does this link help? http://www.mathworks.com/help/toolbox/stats/bsw6a5x.html#bsw6bqq If you are concerned with a specific...

14 years ago | 0

Answered
Visualizing a tree
If you work in 11a or later and if you have Statistics Toolbox, you can boost trees using fitensemble function. Then you can use...

14 years ago | 0

Answered
ClassificationTree with unequal costs
Yes, ClassificationTree always predicts class labels based on posterior probabilities. In that, ClassificationTree/predict devia...

14 years ago | 0

Answered
Classregtree - implementing an additional constraint
If you work in release 9b or later, the function searching for the optimal split in a tree node is coded in C for speed. The sou...

14 years ago | 0

Answered
KERNEL : mean integrated squared error- Bandwidth Selection
The "right" width depends on your assumptions about the fitted distribution. MATLAB does not choose the bandwidth "randomly". It...

14 years ago | 0

| accepted

Answered
How to make matrix a possitive definite
If you have 10935 predictors, your entire training and test data are not going to be enough to make the covariance matrix non-si...

14 years ago | 0

Answered
Leraning classification with most training samples in one category
The answer depends on how you define a "fair" classifier. If the ultimate goal of your analysis is to minimize the overall class...

14 years ago | 2

| accepted

Answered
Difference between individual and cumulative oobMargin of TreeBagger
When you ask for an OOB margin from one tree, you get zero if this observation was in bag for this tree. The margin is undefined...

14 years ago | 0

| accepted