Socially Relevant Venue Clustering from Check-in Data

Abstract

The recent proliferation of location-based social network services has resulted in an abundance of spatial-temporal data on user mobility. Understanding individual and collective mobility patterns is important for many applications. In this study, we examine the similarity of users based on the venues they have visited in the past. In contrast to the previous approaches that measure user similarity based on co-location patterns, here we rst cluster venues in some latent (lower-dimensional) space, which allows us to capture the similarity between two users who have not necessarily visited the exact same venues in the past. We validate our approach on real-world data and demonstrate an improved performance over previous methods.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 12 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.