Can the Ambiance of a Place be Determined by the User Profiles of the People Who Visit It?

Abstract

Recently, several online social networking sites (OSNs; e.g., FourSquare) have emerged that allow people to record and broadcast their presence at physical locations. We examined the viability of using images of frequent patrons listed on an establishment’s FourSquare page to infer the ambiance of the establishment. In Study 1, 10 judges independently rated 25 bars and 24 cafes in Austin, Texas. Ratings were based on the profile pictures of 25 randomly selected persons who frequented each establishment. Ratings tapped ambient physical and psychological qualities (e.g., loud, creepy), typical-patron personalities (e.g., extraverted), and likely patron activities (e.g., dancing). Results indicated considerable inter-judge consensus across the attributes rated (mean ICCs averaged .32, .69. and .33 for the ambiance, patron-personality, and patron-activity variables respectively). In Study 2, a second team of 10 observers visited each target establishment during business hours and rated each location on the same set of variables. These ratings also showed strong inter-judge consensus (mean ICCs averaged .69, .79. and .62 for the ambiance, patron-personality, and patron-activity variables respectively). Correlations between the ratings obtained in Studies 1 and 2 showed surprisingly strong convergence suggesting that the profile photos conveyed valid information about the target establishments.

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 — 27 citations exist upstream.

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