A personalized point-of-interest recommendation model via fusion of geo-social information

摘要: Recently, as location-based social networks (LBSNs) rapidly grow, general users utilize point-of-interest recommender systems to discover attractive locations. Most existing POI recommendation algorithms always employ the check-in data and rich contextual information (e.g., geographical network information) of learn their preference on POIs. Unfortunately, these studies generally suffer from two major limitations: (1) when modeling influence, personalized behavior differences are ignored; (2) implicit influence is seldom exploited. In this paper, we propose a novel approach called GeoEISo. GeoEISo achieves three key goals in work. We develop kernel estimation method with self-adaptive bandwidth model between use Gaussian radial basis function based support vector regression (SVR) predict explicit trust values users, then devise trust-based simultaneously incorporate both into process recommendation. (3) unified geo-social framework which combines well correlations. Experimental results real-world datasets collected Foursquare show that provides significantly superior performances compared other state-of-the-art models.