Try support vector machine: http://svmlight.joachims.org/

This one is from Matlab help (Machine Learning->Supervised Learning->SVM->Functions->fitcsvm->Examples): 

Detect Outliers Using SVM and One-Class Learning

Load Fisher's iris data set. Remove the petal lengths and widths. Treat all irises as coming from the same class.

load fisheriris

X = meas(:,1:2);

y = ones(size(X,1),1);

.... and there are two more steps you need to do.

There are several methods available. if your data univariate you can use the given following

1. Grubbs test

2. Z Scores or Modified Z-Scores

3. Median Absolute Deviation (MADe)

4. Hample Method

6. Generalised ESD

7. Tukey Method

8. Boxplot

9. Quartile Methods

10.Dixon’s Test (The test has some

limitations to n

thanks very much