AUC-MF: Point of Interest Recommendation with AUC Maximization

Han, Peng; Shang, Shuo; Sun, Aixin; Zhao, Peilin; Zheng, Kai; Kalnis, Panos

AUC-MF: Point of Interest Recommendation with AUC Maximization

Files

P17.pdf (948.07 KB)

Type

Poster

Authors

Han, Peng
Shang, Shuo
Sun, Aixin
Zhao, Peilin
Zheng, Kai
Kalnis, Panos

Date

2019-01-13

Abstract

AUC-MF: Point of Interest Recommendation with AUC Maximization Location-based social networks (LSBNs) allow users to check in and share their experiences when they visit a point of interest (POI), such as a museum or a restaurant. With the development and popularity of various LSBN (Fig. 1) platforms e.g., BrightKite, Foursquare, and Gowalla, user check-in data is growing at an unprecedented pace. For instance, Foursquare had more than 50 million active users and more than 8 billion check-ins made by 2016. The availability of abundant amount of user check-in data, enables many studies on recommender systems to further enhance user experiences. POI recommendation aims at finding unvisited locations that a user may be interested in, by learning from users check-in history and other related factors. POI recommendation is challenging for many reasons. One of the most important reasons is that user check-in data is extremely sparse .