Classification of ride-sharing partners based on multiple constraints

This paper describes the formulation of algorithms for measuring the closeness of a match between pairs of potential partners in an online ride-sharing or carpooling system. These algorithms compare realistic routes used by the participants, as well as their schedules. The paper goes on to integrate these measures together with other filtering constraints based on user preferences such as type of carpooling partner, creating a single overall measure that can be used to rank potential matches to a given user of the system.