The explicitly observed social relations from online social platforms have been widely incorporated into recommender systems to mitigate the data sparsity issue. However, the direct usage of explicit social relations may lead to an inferior performance due to the unreliability (e.g., noises) of observed links. To this end, the discovery of reliable relations among users plays a central role in advancing social recommendation. In this paper, we propose a novel approach to adaptively identify implicit friends toward discovering more credible user relations. Particularly, implicit friends are those who share similar tastes but could be distant from each other on the network topology of social relations. Methodologically, to find the implicit friends for each user, we first model the whole system as a heterogeneous information network, and then capture the similarity of users through the meta-path based embedding representation learning. Finally, based on the intuition that social relations have varying degrees of impact on different users, our approach adaptively incorporates different numbers of similar users as implicit friends for each user to alleviate the adverse impact of unreliable social relations for a more effective recommendation. Experimental analysis on three real-world datasets demonstrates the superiority of our method and explain why implicit friends are helpful in improving social recommendation.

Recently, personalised search engines and recommendation systems have been widely adopted by users who require assistance in searching, classifying, and filtering information. This paper presents an overview of the field of personalisation systems and describes current state-of-the-art methods and techniques. It reviews approaches for (1) user profiling, including behaviour, preference, and intention modelling; (2) content modelling, comprising content representation, analysis, and classification; and (3) filtering methods for recommendation, classified into four main categories: rule-based, contentbased, collaborative, and hybrid filtering. The paper also discusses personalisation systems in different domains, and various techniques and their limitations. Finally, it identifies several issues and possible directions for further research that can improve recommendation capabilities and enhance personalised systems.

Recent reports from industry show that social recommender systems consistently fail in practice. According to the negative findings, the failure is attributed to: (1) a majority of users only have a very limited number of neighbors in social networks and can hardly benefit from relations; (2) social relations are noisy but they are often indiscriminately used; (3) social relations are assumed to be universally applicable to multiple scenarios while they are actually multi-faceted and show heterogeneous strengths in different scenarios. Most existing social recommendation models only consider the homophily in social networks and neglect these drawbacks.In this paper we propose a deep adversarial framework based on graph convolutional networks (GCN) to address these problems. Concretely, for the relation sparsity and noises problems, a GCN-based autoencoder is developed to augment the relation data by encoding high-order and complex connectivity patterns, and meanwhile is optimized subject to the constraint of reconstructing the original social profile to guarantee the validity of new identified neighborhood. After obtaining enough purified social relations for each user, a GCN-based attentive social recommendation module is designed to capture the heterogeneous strengths of social relations. These designs deal with the three problems faced by social recommender systems respectively. Finally, we adopt adversarial training to unify and intensify all components by playing a minimax game and ensure a coordinated effort to enhance social recommendation. Experimental results on multiple open datasets demonstrate the superiority of our framework and the ablation study confirms the importance and effectiveness of each component.

a b s t r a c tDue to the open nature of recommender systems, collaborative recommender systems are vulnerable to profile injection attacks, in which malicious users inject attack profiles into the rating matrix in order to bias the systems' ranking list. Recommender systems are highly vulnerable to shilling attacks, both by individuals and groups. Most of previous research focuses only on the differences between genuine profiles and attack profiles, ignoring the group characteristics in attack profiles of an attack. There also exist class unbalance problems in supervised detecting methods, the detecting performance is not as good when the amount of samples of attack profiles in training set is smaller. In this paper, we study the use of SVM based method and group characteristics in attack profiles. A two phase detecting method SVM-TIA is proposed based on these two methods. In the first phase, Borderline-SMOTE method is used to alleviate the class unbalance problem in classification; a rough detecting result is obtained in this phase; the second phase is a fine-tuning phase whereby the target items in the potential attack profiles set are analyzed. We conduct tests on the MovieLens 100 K Dataset and compare the performance of SVM-TIA with other shilling detecting methods to demonstrate the effectiveness of the proposed approach.

Self-supervised learning (SSL), which can automatically generate ground-truth samples from raw data, holds vast potential to improve recommender systems. Most existing SSL-based methods perturb the raw data graph with uniform node/edge dropout to generate new data views and then conduct the self-discrimination based contrastive learning over different views to learn generalizable representations. Under this scheme, only a bijective mapping is built between nodes in two different views, which means that the self-supervision signals from other nodes are being neglected. Due to the widely observed homophily in recommender systems, we argue that the supervisory signals from other nodes are also highly likely to benefit the representation learning for recommendation. To capture these signals, a general socially-aware SSL framework that integrates tri-training is proposed in this paper. Technically, our framework first augments the user data views with the user social information. And then under the regime of tri-training for multi-view encoding, the framework builds three graph encoders (one for recommendation) upon the augmented views and iteratively improves each encoder with self-supervision signals from other users, generated by the other two encoders. Since the tri-training operates on the augmented views of the same data sources for self-supervision signals, we name it self-supervised tri-training. Extensive experiments on multiple real-world datasets consistently validate the effectiveness of the self-supervised tritraining framework for improving recommendation.

No abstract

Web service recommendation is one of the key problems in service computing, especially in the case of a large number of service candidates. The QoS (quality of service) values are usually leveraged to recommend services that best satisfy a user’s demand. There are many existing methods using collaborative filtering (CF) to predict QoS missing values, but very limited works can leverage the network location information in the user side and service side. In real-world service invocation scenario, the network location of a user or a service makes great impact on QoS. In this paper, we propose a novel collaborative recommendation framework containing three novel prediction models, which are based on two techniques, i.e. matrix factorization (MF) and network location-aware neighbor selection. We first propose two individual models that have the capability of using the user and service information, respectively. Then we propose a unified model that combines the results of the two individual models. We conduct sufficient experiments on a real-world dataset. The experimental results demonstrate that our models achieve higher prediction accuracy than baseline models, and are not sensitive to the parameters.

Most of the recent studies of social recommendation assume that people share similar preferences with their friends and the online social relations are helpful in improving traditional recommender systems. However, this assumption is often untenable as the online social networks are quite sparse and a majority of users only have a small number of friends. Besides, explicit friends may not share similar interests because of the randomness in the process of building social networks. Therefore, discovering a number of reliable friends for each user plays an important role in advancing social recommendation. Unlike other studies which focus on extracting valuable explicit social links, our work pays attention to identifying reliable friends in both the observed and unobserved social networks. Concretely, in this paper, we propose an end-to-end social recommendation framework based on Generative Adversarial Nets (GAN). The framework is composed of two blocks: a generator that is used to produce friends that can possibly enhance the social recommendation model, and a discriminator that is responsible for assessing these generated friends and ranking the items according to both the current user and her friends' preferences. With the competition between the generator and the discriminator, our framework can dynamically and adaptively generate reliable friends who can perfectly predict the current user' preference at a specific time. As a result, the sparsity and unreliability problems of explicit social relations can be mitigated and the social recommendation performance is significantly improved. Experimental studies on real-world datasets demonstrate the superiority of our framework and verify the positive effects of the generated reliable friends.