Machine Learning project: Reality Mining.

One of the most challenging areas for the application of machine learning algorithms is Reality Mining [1], in particular the understanding of social dynamics from the analysis of data describing social interactions (e.g., data about mobile communications, coworking interactions, information exchange, computer use, etc.). Several recent projects have proposed experiments in which this type of data is collected [2, 3, 1, 4, 5] and a number of approaches that analyze this data have been also published [6, 7, 8, 9, 10, 11]. However, the question of exploring new and creative applications of machine learning in this domain is relevant. In this project the students will use the Reality Mining Dataset http://realitycommons. media.mit.edu/realitymining.html [1] which contains mobile data from 75 students or faculty in the MIT Media Laboratory, and 25 incoming students at the MIT Sloan business school adjacent to the Media Laboratory. The goal of the experiment was to explore the capabilities of the smart phones to enable social scientists to investigate human interactions.