A New Unsupervised Methodology of Descriptive Text Mining for Knowledge Graph Learning

Rare diseases pose particular challenges to patients, families, caregivers, clinicians and researchers. Currently, more than 6000 rare diseases are described (but up to 7000 are estimated) and more than 350 million people live with them (5\% of the world population). Due to the scarce availability of information and their disintegration, in recent years we are witnessing strong growth of patient communities on social platforms such as Facebook. The work presented in this thesis is intended to extract knowledge from the large availability of unstructured text generated by the users over time, in order to represent it in an organized way and to make logical reasoning above. Starting from the awareness of the need to integrate different methodologies in complex domains, the research shows a combined use of Text Mining and Semantic Web techniques, taking Esophageal Achalasia as a case study. In particular, an ontology is created to extend ORDO and introduce a patient-centered vision into the world of linked data. The significance of this development is that it potentially constitutes the basis of a project that can allow rapid access to many high-value information (in topics such as symptomatology, epidemiology, diagnosis, treatments, drugs, nutrition, lifestyle), responding to patients' questions and providing them with an additional tool for decision making, minimizing costs through the automatic retrieval of these data and increasing the productivity of investigators.