This document aims to show the relevance of Big Data as a tool for improving urban mobility management. To this end, mobility data from the STI (Integrated Tariff System) ambit of the Barcelona area are represented and analysed, and the factor that most determines the choice of transport mode of the inhabitants of the AMB (Barcelona Metropolitan Area) in their journeys to Barcelona is obtained using machine learning techniques. The data representation and machine learning exercise has been done using a virtual environment called Google Colaboratory (also called Google Colab for short), which allows you to run and program Phyton in the browser without previous configuration and with free access to GPUs (Graphics Processing Units). The structure of the document is as follows: A first chapter dedicated to characterize the AMB (municipalities that compose it, location, demography and economy). Then, data from the STI ambit of the Barcelona area are represented and analyzed. The third chapter is a tutorial about the basic functioning of the Google Colab environment. Finally, a predictive analysis is made in order to know what is the main cause (distance, cost of the trip in private transport or cost of the trip in public transport) that motivates the inhabitants of the AMB to choose one mode of transport or another in their trips to Barcelona.