Implementing an adaptive boosting algorithm (AdaBoost) in MatLab using decision stumps learned using information gain as the weak learners.
I have used AdaBoost to classify the "notorious" handwritten digits problem described in the machine learning book Learning From Data by Yaser S. Abu-Mostafa, Malik Magdon-Ismail, and Hsuan-Tien Lin.
The training and testing sets can be found here. The test set is notoriously "difficult", and a 2.5% error rate is excellent.