In the last few years, the way people share information has changed a lot due to social networks. They have become one of the principal means to share emotions and feelings. In Spain, as in other countries where football is the first sport in terms of the number of followers, one of the topics that is frequently discussed in social networks is football, especially during the match time. Nowadays, Twitter has become an interesting scenario for social analysis. Twitter con- sists in a micro-blogging service and can be used to find about events and news in real time from anywhere in the world. As Twitter posts (tweets) are short and are being generated constantly, they are well-suited sources of streaming data. In this project we are going to focus on the way people express emotions during football matches. The main objective of the project is the development of a Machine Learning system, which will be able to classify football related tweets according to the emotion that they express. As a secondary objective, we will develop a football corpus with an annotation Crowdsourcing platform to classify and analyze the tweets. To carry out the project, we are going to develop a corpus using capture techniques in Twitter and the python programming language. Firstly, we will make the annotation and classification of the tweets with the platform of Crowdsourcing Pybossa. Later, for the development of the classifier, we are going to use Automatic learning tools such as scikit-learn with Supervised Automatic Language techniques and Natural Language Processing tools (NLP). Finally, we will present the service in a graphical display with a dashboard.