Properties of the Mallows Model Depending on the Number of Alternatives: A Warning for an Experimentalist

Speaker(s)

Sonja Kraiczy

Affiliation

University of Oxford

Date

March 2, 2023, noon

Room

room 4050

Seminar

Seminar Games, Mechanisms, and Social Networks

---

The Mallows model is a popular distribution for ranked data. We empirically and theoretically analyze how the properties of rankings sampled from the Mallows model change when increasing the number of alternatives. We find that real-world data behaves differently than the Mallows model, yet is in line with its recent variant proposed by Boehmer et al. [IJCAI ’21]. As part of our study, we issue several warnings about using the model.

---