Gender gap in scientific studies: a data mining approach

In the present work we use clustering techniques to study seven cohorts of students belonging to scientific degree courses sharing the same self assessment test required to students before enrolling in the University of Florence. The work focuses on the study of gender differences, in terms of the number of enrollments in the scientific degree programs and of the students productivity during the first year. The goal is to use data mining techniques to give analytical evidence that the result in scientific studies does not depend on gender and then use these results in tutoring activities to encourage girls enrollmen.