Exploiting Geo-Social Correlations to Improve Pairwise Ranking for Point-of-Interest Recommendation

Recently, as location-based social network(LBSN) rapidly grow, point-of-interest(POI) recommendation has become an important way to help people locate interesting places. Nowadays, there have been deep studies conducted on the geographical and social influence in the point-of-interest recommendation model based on the rating prediction. The fact is, however, relying solely on the rating fails to reflect the user's preferences very accurately, because the users are most concerned with the list of ranked point-of-interests(POIs) on the actual output of recommender systems. In this paper, we propose a co-pairwise ranking model called Geo-Social Bayesian Personalized Ranking model(GSBPR), which is based on the pairwise ranking with the exploiting geo-social correlations by incorporating the method of ranking learning into the process of POI recommendation. In this model, we develop a novel BPR pairwise ranking assumption by injecting users' geo-social preference. Based on this assumption, the POI recommendation model is reformulated by a three-level joint pairwise ranking model. And the experimental results based on real datasets show that the proposed method in this paper enjoys better recommendation performance compared to other state-of-the-art POI recommendation models.

Recommending Personalized POIs from Location Based Social Network

Location based social networks（ LBSNs） provide location specific data generated from smart phone into online social networks thus people can share their points of interest（ POIs）. POI collections are complex and can be influenced by various factors,such as user preferences,social relationships and geographical influence. Therefore,recommending new locations in LBSNs requires to take all these factors into consideration. However,one problem is how to determine optimal weights of influencing factors in an algorithm in which these factors are combined. The user similarity can be obtained from the user check-in data,or from the user friend information,or based on the different geographical influences on each user＇s check-in activities. In this paper,we propose an algorithm that calculates the user similarity based on check-in records and social relationships,using a proposed weighting function to adjust the weights of these two kinds of similarities based on the geographical distance between users. In addition,a non-parametric density estimation method is applied to predict the unique geographical influence on each user by getting the density probability plot of the distance between every pair of user＇s check-in locations. Experimental results,using foursquare datasets,have shown that comparisons between the proposed algorithm and the other five baseline recommendation algorithms in LBSNs demonstrate that our proposed algorithm is superior in accuracy and recall,furthermore solving the sparsity problem.

Social Trust Aware Item Recommendation for Implicit Feedback

Social trust aware recommender systems have been well studied in recent years. However, most of existing methods focus on the recommendation scenarios where users can provide explicit feedback to items. But in most cases, the feedback is not explicit but implicit. Moreover, most of trust aware methods assume the trust relationships among users are single and homogeneous, whereas trust as a social concept is intrinsically multi-faceted and heterogeneous. Simply exploiting the raw values of trust relations cannot get satisfactory results. Based on the above observations, we propose to learn a trust aware personalized ranking method with multi-faceted trust relations for implicit feedback. Specifically, we first introduce the social trust assumption -- a user＇s taste is close to the neighbors he/she trusts into the Bayesian Personalized Ranking model. To explore the impact of users＇ multi-faceted trust relations, we further propose a category- sensitive random walk method CRWR to infer the true trust value on each trust link. Finally, we arrive at our trust strength aware item recommendation method SocialBPRcawn by replacing the raw binary trust matrix with the derived real-valued trust strength. Data analysis and experimental results on two real-world datasets demonstrate the existence of social trust influence and the effectiveness of our social based ranking method SocialBPRcawR in terms of AUC （area under the receiver operating characteristic curve）.

Exploiting Pre-Trained Network Embeddings for Recommendations in Social Networks

Recommender systems as one of the most efficient information filtering techniques have been widely studied in recent years. However, traditional recommender systems only utilize user-item rating matrix for recommendations, and the social connections and item sequential patterns are ignored. But in our real life, we always turn to our friends for recommendations, and often select the items that have similar sequential patterns. In order to overcome these challenges, many studies have taken social connections and sequential information into account to enhance recommender systems. Although these existing studies have achieved good results, most of them regard social influence and sequential information as regularization terms, and the deep structure hidden in social networks and rating patterns has not been fully explored. On the other hand, neural network based embedding methods have shown their power in many recommendation tasks with their ability to extract high-level representations from raw data. Motivated by the above observations, we take the advantage of network embedding techniques and propose an embedding-based recommendation method, which is composed of the embedding model and the collaborative filtering model. Specifically, to exploit the deep structure hidden in social networks and rating patterns, a neural network based embedding model is first pre-trained, where the external user and item representations are extracted. Then, we incorporate these extracted factors into a collaborative filtering model by fusing them with latent factors linearly, where our method not only can leverage the external information to enhance recommendation, but also can exploit the advantage of collaborative filtering techniques. Experimental results on two real-world datasets demonstrate the effectiveness of our proposed method and the importance of these external extracted factors.

Textual-geographical-social aware point-of-interest recommendation

The rapid development of location-based social networks（LBSNs） has provided an unprecedented opportunity for better location-based services through point-of-interest（POI） recommendation. POI recommendation is personalized, location-aware, and context depended. However, extreme sparsity of user-POI matrix creates a severe challenge. In this paper we propose a textual-geographical-social aware probabilistic matrix factorization method for POI recommendation. Our model is textual-geographical-social aware probabilistic matrix factorization called TGS-PMF, it exploits textual information, geographical information, social information, and incorporates these factors effectively. First, we exploit an aggregated latent Dirichlet allocation（LDA） model to learn the interest topics of users and infer the interest POIs by mining textual information associated with POIs and generate interest relevance score. Second, we propose a kernel estimation method with an adaptive bandwidth to model the geographical correlations and generate geographical relevance score. Third, we build social relevance through the power-law distribution of user social relations to generate social relevance score. Then, our exploit probabilistic matrix factorization model（PMF） to integrate the interest, geographical, social relevance scores for POI recommendation. Finally, we implement experiments on a real LBSN check-in dataset. Experimental results show that TGS-PMF achieves significantly superior recommendation quality compared to other state-of-the-art POI recommendation techniques.

Two-sided regularization model based on probabilistic matrix factorization and quantum similarity for recommender systems

Nowadays,with the advent of the age of Web 2.0,several social recommendation methods that use social network information have been proposed and achieved distinct developments.However,the most critical challenges for the existing majority of these methods are:(1)They tend to utilize only the available social relation between users and deal just with the cold-start user issue.(2)Besides,these methods are suffering from the lack of exploitation of content information such as social tagging,which can provide various sources to extract the item information to overcome the cold-start item and improve the recommendation quality.In this paper,we investigated the efficiency of data fusion by integrating multi-source of information.First,two essential factors,user-side information,and item-side information,are identified.Second,we developed a novel social recommendation model called Two-Sided Regularization(TSR),which is based on the probabilistic matrix factorization method.Finally,the effective quantum-based similarity method is adapted to measure the similarity between users and between items into the proposed model.Experimental results on the real dataset show that our proposed model TSR addresses both of cold-start user and item issues and outperforms state-ofthe-art recommendation methods.These results indicate the importance of incorporating various sources of information in the recommendation process.