Fused matrix factorization with geographical and social influence in location-based social networks
Citations
731 citations
Cites background from "Fused matrix factorization with geo..."
...There exists other work that incorporates social link information into POI recommendations, such as the probabilistic generative model-based method [21], and matrix factorization-based method [4]....
[...]
...The focus of [4, 21] is to explore social link information for POI recommendations and their problem setting is different from ours....
[...]
687 citations
Cites methods from "Fused matrix factorization with geo..."
...2011) or the multi-center gaussian model (Cheng et al. 2012)....
[...]
...And for spatial information, the distance between locations is calculated and the prediction is made based on power law distribution (Ye et al. 2011) or the multi-center gaussian model (Cheng et al. 2012)....
[...]
582 citations
Cites background or methods from "Fused matrix factorization with geo..."
...Concentrating on modeling the distance distribution may ignore the multi-center characteristics of individual visiting locations according to [2]....
[...]
...This algorithm has been exploited in [2, 12] for POI recommendation....
[...]
...cal information to assist POI recommendation [24, 2, 12, 26] by modeling the well-known spatial clustering phenomenon, these approaches are almost independent of the procedure for collaborative filtering, particularly, matrix factorization....
[...]
...To avoid the cost in computing the distance between paired locations, in [2, 12], the authors modeled the spatial clustering phenomenon in terms of geo-clustering and tried to estimate individual spatial distribution....
[...]
..., B-NMF ) improves compared to MF-Freq since it can model the skewness of visit frequency [2, 12]....
[...]
548 citations
Cites background from "Fused matrix factorization with geo..."
...Some previous works [12], [25], [26] have studied the probability of location visiting w....
[...]
522 citations
Cites background or methods or result from "Fused matrix factorization with geo..."
...The other line of work focuses on LBSN data, which is very sparse and large-scale [Ye et al., 2010; 2011; Cheng et al., 2012]....
[...]
...…e.g., modeling the check-in probability to the distance of the whole check-in history by power-law distribution [Ye et al., 2011], modeling users’ multi-center check-in behaviors via multicenter Gaussians [Cheng et al., 2012], and etc., have been addressed and fused with traditional CF algorithms....
[...]
..., 2011], modeling users’ multi-center check-in behaviors via multicenter Gaussians [Cheng et al., 2012], and etc....
[...]
..., 2011b] and the Gowalla data from [Cheng et al., 2012]....
[...]
...Currently, geographical influence, e.g., modeling the check-in probability to the distance of the whole check-in history by power-law distribution [Ye et al., 2011], modeling users’ multi-center check-in behaviors via multicenter Gaussians [Cheng et al., 2012], and etc., have been addressed and fused with traditional CF algorithms....
[...]
References
4,022 citations
"Fused matrix factorization with geo..." refers background or methods in this paper
...PMF: this is a well-known method in matrix factorization (Salakhutdinov and Mnih 2007)....
[...]
...Probabilistic Matrix Factorization (PMF) PMF is one of the most famous MF models in collaborative filtering (Salakhutdinov and Mnih 2007)....
[...]
...Matrix Factorization (MF) is one of the most popular methods for recommender systems (Salakhutdinov and Mnih 2007; 2008; Bell, Koren, and Volinsky 2007; Koren 2009)....
[...]
...The training time for the matrix factorization models scales linearly with the number of observations (Salakhutdinov and Mnih 2007; Ma et al. 2011b)....
[...]
2,922 citations
"Fused matrix factorization with geo..." refers background or methods or result in this paper
...This indicates that less than 10% of a user’s check-ins are also visited by his/her friends, which is similar to the statistic reported in (Cho, Myers, and Leskovec 2011)....
[...]
...In addition, our statistic is also a little different from the two states (“home” and “office”) check-in behavior mentioned in (Cho, Myers, and Leskovec 2011)....
[...]
...The statistics are a little different from those in (Cho, Myers, and Leskovec 2011), but the overall trend is similar....
[...]
...This means that users usually visit several important places, e.g., home, office,and some stores or bars, with very high frequency, while most of other places are seldom visited....
[...]
1,903 citations
1,621 citations
"Fused matrix factorization with geo..." refers methods in this paper
...To solve large-scale recommendation problems, matrix factorization is a promising tool due to its success in Netflix competition (Bell, Koren, and Volinsky 2007; Koren 2009)....
[...]
...To solve large-scale recommendation problems, matrix factorizati on is a promising tool due to its success in Netflix competition (Bell, Koren, and Volinsky 2007; Koren 2009)....
[...]
...Matrix Factorization (MF) is one of the most popular methods for recommender systems (Salakhutdinov and Mnih 2007; 2008; Bell, Koren, and Volinsky 2007; Koren 2009)....
[...]
1,573 citations
"Fused matrix factorization with geo..." refers background or methods in this paper
...We adopts the PMF with Social Regularization (PMFSR) (Ma et al. 2011b), whose objective function is defined as follows: min U,L Ω(U,L) = |U|∑ i=1 |L|∑ j=1 Iij(g(Fij)− g(U T i Lj)) 2 + β |U|∑ i=1 ∑ f∈F(i) Sim(i, f)‖Ui − Uf‖ 2 F + λ1‖U‖ 2 F + λ2‖L‖ 2 F , (4) whereF(i) is the set of friends for…...
[...]
...We adopts the PMF with Social Regularization (PMFSR) (Ma et al. 2011b), whose objective function is defined as follows:...
[...]
...PMF with Social Regularization (PMFSR): this method is proposed to include the social friendship under the PMF framework (Ma et al. 2011b)....
[...]
...• PMFSR attains a little better results than those of PMF....
[...]
...4), we turn to Probabilistic Factor Models (PFM) (Chen et al. 2009; Ma et al. 2011a), which can model the frequency data directly....
[...]