In this chapter, we explored the possibility of detecting fraudulent transactions using various ensemble learning methods. While some performed better than others, due to the dataset's nature, it is difficult to produce good results without resampling the dataset in some way (either over-sampling or under-sampling).

We were able to show how to use each ensemble learning method and how to explore the possibility of fine-tuning its respective parameters in order to achieve better performance. In the next chapter, we will try to leverage ensemble learning techniques in order to predict Bitcoin prices.