Practical Machine Learning and Image Processing: For Facial Recognition, Object Detection, and Pattern Recognition Using Python

Appendix: Important Concepts and Terminology

Adaboost

Adaboost is a technique that uses multiple classification algorithms throughout the training data. First we take a random sample from the dataset and call it a training set . Adaboost assigns weights to each and every example in the training set. After that, the first classifier is run. When the classification is done, Adaboost looks at the misclassified examples in the training set. They are then assigned greater weights, as are the correctly classified examples. This happens so that when the second classifier is run, it takes into consideration examples with greater weights.

In addition, weights are assigned to classifiers as well. When the classifier gives an output, its error is calculated. ...