The thesis will focus on the use of NLP methods and Machine Learning for research in the humanities and social sciences (SHS). One particular interest is focused on the study of the added value that these tools offer to research in SHS and in return how questions typical of SHS can feed the epistemology of computer science. Today, we have a unique opportunity to revisit the history of sciences by leveraging recent and powerful automatized tools, such as the ones provided by Machine Learning and natural language processing. In this thesis, we aim at proposing new machine learning techniques for analyzing computer science texts.