Ensemble Methods for Machine Learning

3 Heterogeneous parallel ensembles: Combining strong learners

This chapter covers

• Combining base-learning models by performance-based weighting
• Combining base-learning models with meta-learning by stacking and blending
• Avoiding overfitting by ensembling with cross validation
• Exploring a large-scale, real-world, text-mining case study with heterogeneous ensembles

In the previous chapter, we introduced two parallel ensemble methods: bagging and random forests. These methods (and their variants) train homogeneous ensembles, where every base estimator is trained using the same base-learning algorithm. For example, in bagging classification, all the base estimators are decision-tree classifiers.

In this chapter, we continue exploring parallel ensemble ...