Modelling urban bus fleet emissions with machine learning boosting methods: City of Madrid

Boosting is a machine learning methodology which consists in an ensemble (set) of similar models estimated from the same data set. It is an iterative and cumulative algorithm intended to minimize the error of a single "weak" model. The purpose of this work is to assess the applicability of this technique to the modelling and prediction of instantaneous emissions of urban buses in the city of Madrid.