Enhancing Collaborative Filtering with Friendship Information

2013

Two main approaches to using social network information in recommendation have emerged: augmenting collaborative filtering with social data and algorithms that use only ego-centric data. We compare the two approaches using movie and music data from Facebook, and hashtag data from Twitter. We find that recommendation algorithms based only on friends perform no worse than those based on the full network, even though they require much less data and computational resources. Further, our evidence suggests that locality of preference, or the non-random distribution of item preferences in a social network, is a driving force behind the value of incorporating social network information into recommender algorithms. When locality is high, as in Twitter data, simple k-nn recommenders do better based only on friends than they do if they draw from the entire network. These results help us understand when, and why, social network information is likely to support recommendation systems, and show that systems that see ego-centric slices of a complete network (such as websites that use Facebook logins) or have computational limitations (such as mobile devices) may profitably use ego-centric recommendation algorithms.

The current industry standard for recommender system uses variants of collaborative filtering (CF), where recipient-source relationships are determined by the extent to which the recipient and source share interests. This research attempts to improve the performance of these CF recommender systems by identifying additional measures of relationship indicators based on theories from communication and marketing. We developed a social filtering model that incorporates these various social measures (e.g. Trust, Reputation, Interaction Frequency, and Relationship Duration), and conducted an empirical study to test the model. The results from the study show small, but significant, improvements for various social relationships. We plan to build on these preliminary results to further consolidate our research on using social relationship in recommender systems.

The existing social networking providers advocate close friends to help end users according to their own interpersonal charts, which most likely are not the most likely to help reflect a user's personal preferences about pal assortment throughout real life. Within this cardstock, all of us existing Friendsbook, a book semantic primarily based pal advice technique for internet sites, which recommends close friends to help end users according to their own way of life rather than interpersonal charts. Through benefiting from sensor-rich smartphones, Friendsbook detects way of life involving end users through user-centric sensor information, steps your likeness involving way of life between end users, and also recommends close friends to help end users when their own way of life include large likeness. Motivated by simply textual content exploration, all of us style a user's daily life while lifestyle files, from which his/her way of life are generally produced with the Latent Dirichlet Algorithm protocol. Most of us more recommend a likeness metric to help gauge your likeness involving way of life between end users, and also estimate users' result with regard to way of life having a friend-matching chart. When receiving a ask, Friendsbook earnings a summary of those with greatest advice results for the dilemma person. Eventually, Friensdbook integrates a opinions procedure for boosting your advice precision. We now have carried out Friendsbook for the Android-based smartphones, and also looked at its efficiency about both equally small-scale studies and also large-scale simulations. The final results indicate that the suggestions accurately reflect your personal preferences involving end users throughout picking close friends.

2013

In the Internet era, online social media emerged as the main tool for sharing opinions and information among individuals. In this work we study an adaptive model of a social network where directed links connect users with similar tastes, and over which information propagates through social recommendation. Agent-based simulations of two different artificial settings for modeling user tastes are compared with patterns seen in real data, suggesting that users differing in their scope of interests is a more realistic assumption than users differing only in their particular interests. We further introduce an extensive set of similarity metrics based on users' past assessments, and evaluate their use in the given social recommendation model with both artificial simulations and real data. Superior recommendation performance is observed for similarity metrics that give preference to users with small scope-who thus act as selective filters in social recommendation.

With the advent and popularity of social network, more and more users like to share their experiences, such as ratings, reviews, and blogs. The new factors of social network like interpersonal influence and interest based on circles of friends bring opportunities and challenges for recommender system (RS) to solve the cold start and sparsity problem of datasets. Some of the social factors have been used in RS, but have not been fully considered. In this paper, three social factors, personal interest, interpersonal interest similarity, and interpersonal influence, fuse into a unified personalized recommendation model based on probabilistic matrix factorization. The factor of personal interest can make the RS recommend items to meet users' individualities, especially for experienced users. Moreover, for cold start users, the interpersonal interest similarity and interpersonal influence can enhance the intrinsic link among features in the latent space. We conduct a series of experiments on three rating datasets: Yelp, MovieLens, and Douban Movie. Experimental results show the proposed approach outperforms the existing RS approaches.

Springer eBooks, 2012

Social networks provide users with information about their friends, their activities, and their preferences. In this paper we study the effectiveness of movie recommendations computed from such communicated preferences. We present a set of social movie recommendation algorithms, which we implemented on top of the Facebook social network, and we compare their effectiveness in influencing user decisions. We also study the effect of showing users a justification for the recommendations, in the form of the profile pictures of the friends that caused the recommendation. We show that social movie recommendations are generally accurate. Furthermore, 80% of the users that are undecided on whether to accept a recommendation are able to reach a decision upon learning of the identities of the users behind the recommendation. However, in 27% of the cases, they decide against watching the recommended movies, showing that revealing identities can have a negative effect on recommendation acceptance.

ACM Transactions on the Web, 2014

User-based collaborative filtering systems suggest interesting items to a user relying on similar-minded people called neighbors. The selection and weighting of these neighbors characterize the different recommendation approaches. While standard strategies perform a neighbor selection based on user similarities, trust-aware recommendation algorithms rely on other aspects indicative of user trust and reliability. In this paper we restate the trust-aware recommendation problem, generalizing it in terms of performance prediction techniques, whose goal is to predict the performance of an information retrieval system in response to a particular query. We investigate how to adopt the above generalization to define a unified framework where we conduct an objective analysis of the effectiveness (predictive power) of neighbor scoring functions. The proposed framework enables discriminating whether recommendation performance improvements are caused by the used neighbor scoring functions or by the ways these functions are used in the recommendation computation. We evaluated our approach with several state-ofthe-art and novel neighbor scoring functions on three publicly available datasets. By empirically comparing four neighbor quality metrics and thirteen performance predictors, we found strong predictive power for some of the predictors with respect to certain metrics. This result was then validated by checking the final performance of recommendation strategies where predictors are used for selecting and/or weighting user neighbors. As a result, we have found that, by measuring the predictive power of neighbor performance predictors, we are able to anticipate which predictors are going to perform better in neighbor scoring powered versions of a user-based collaborative filtering algorithm.

… Systems Design and …, 2011

Matrix Factorization techniques have been successfully applied to raise the quality of suggestions generated by Collaborative Filtering Systems (CFSs). Traditional CFSs based on Matrix Factorization operate on the ratings provided by users and have been recently extended to incorporate demographic aspects such as age and gender. In this paper we propose to merge CFS based on Matrix Factorization and information regarding social friendships in order to provide users with more accurate suggestions and rankings on items of their interest. The proposed approach has been evaluated on a real-life online social network; the experimental results show an improvement against existing CFSs. A detailed comparison with related literature is also present.

2009

Social trust relationships between users in social networks speak to the similarity in opinions between the users, both in general and in important nuanced ways. They have been used in the past to make recommendations on the web. New trust metrics allow us to easily cluster users based on trust. In this paper, we investigate the use of trust clusters as a new way of improving recommendations. Previous work on the use of clusters has shown the technique to be relatively unsuccessful, but those clusters were based on similarity rather than trust. Our results show that when trust clusters are integrated into memory-based collaborative filtering algorithms, they lead to statistically significant improvements in accuracy. In this paper we discuss our methods, experiments, results, and potential future applications of the technique.

Information

Recent research in the field of recommender systems focuses on the incorporation of social information into collaborative filtering methods to improve the reliability of recommendations. Social networks enclose valuable data regarding user behavior and connections that can be exploited in this area to infer knowledge about user preferences and social influence. The fact that streaming music platforms have some social functionalities also allows this type of information to be used for music recommendation. In this work, we take advantage of the friendship structure to address a type of recommendation bias derived from the way collaborative filtering methods compute the neighborhood. These methods restrict the rating predictions for a user to the items that have been rated by their nearest neighbors while leaving out other items that might be of his/her interest. This problem is different from the popularity bias caused by the power-law distribution of the item rating frequency (long-...