MentorMatch was developed to facilitate the laborious, time consuming, and emotional task of choosing mentor-mentee pairings based on mutual preference. With just a spreadsheet of ratings, our program turns a two hour process into a two second solution that is inherently free of bias. Mentor match ensures that every person matched receives the best possible pairing.

This program analyzes a CSV that contains data of how potential mentors (bigs) ranked mentees (littles) and vice versa. It assumes that the number of bigs is greater than or equal to the number of littles. Therefore, each little is guaranteed a big. However, each potential big is not guaranteed to be matched with a little. The maximum number of persons that can be preffed is 10, and the minimum is 5. A form for creating the correctly formatted CSV can be found at

Bigs and littles are both stored as instances of a Person class. Big and little objects are stored in two dictionaries respectively, so that their preferences are accessible by their name. A scores matrix totals the mutual preference values, giving added weight to how a little ranked a big. In the case of a tie, matching is determined by how the little ranked the big.

MentorMatch was developed with the Big/Little relationships associated with sorority and fraternity life in mind. However, it can be used for mutually preferenced mentor/mentee pairings in any organization, sports team, or workplace. It allows for an anonymous submission of rankings by members and a fair, mathematical matching process.

Built With

Share this project: