Recommender systems based on social networks

Sun, Zhanquan; Han, Lixin; Huang, Wen; Wang, Xueting; Zeng, Xiaoqin; Wang, Min; Yan, Hong

doi:10.1016/j.jss.2014.09.019

Cited by 157 publications

(128 citation statements)

References 7 publications

Supporting

Mentioning

127

Contrasting

Unclassified

Order By: Relevance

“…The majority of matrix factorization-based approaches reviewed in Table 12.3 belong to this group -SocialMF [54], Social Regularization [81], CircleCon [137], PWS [132], De Meo's [29] and RSboSN [116]. Social Regularization proposed by Ma, et al [81] is a regularization method to consider the tastes of target users' friends differently depending on the similarity with the target users.…”

Section: Regularization Methodsmentioning

confidence: 99%

“…Macedo, et al [82] considered not only users' event attendance records but also various contextual information of events such as topics of events, the locations, and the temporal information. Sun, et al [116] and Lee & Brusilovsky [71,74] used social tags to improve the quality of bookmark-based recommendations. Bonhard, et al [14] took advantage of users' demographic information to recommend movies to watch.…”

Section: Input Data Types Of Social Link-based Recommendationsmentioning

confidence: 99%

“…Sun, et al [116] Matrix Factorization: The authors clustered users' friends into subgroups and also clustered items into subgroups according to user-item bookmark/tag similarities. Then, a subgroup of friends, whose tastes are similar to target users, and a subgroup of items, which…”

Section: Ndcg Ranking Evaluation Metric (@10)mentioning

confidence: 99%

See 2 more Smart Citations

Recommendations Based on Social Links

Lee

Brusilovsky

2018

Social Information Access

View full text Add to dashboard Cite

Abstract. The goal of this chapter is to give an overview of recent works on the development of social link-based recommender systems and to offer insights on related issues and the future direction. Among several branches of social recommendations, specifically, this chapter focuses on recommendations, which are founded on users' self-defined (i.e., explicit) social links and target to suggest desirable items, not social links of interests. This chapter provides a brief overview over the technical needs for social link-based recommendations and the studies explaining the viability of users' social networks as useful information sources from a social science point of view. Then, the in-depth understanding of existing recommendations based on users' social links will be addressed. Lastly, various issues and future directions of social link-based recommendation research are put forward. IntroductionThe remarkable popularity of social media encourages web users to participate in various online activities and to generate data on an unprecedented scale. Given exponentially growing social media data volumes, personalized recommendation technologies have proven the effectiveness as a solution of the information overload problem; have offered positive user experiences with more relevant contents; and have presented competitive advantage to the social media business [119]. As predictive analysis of online users' tastes, recommendation technology adaptively filters out unnecessary data and selectively chooses presumably favorable information according to users' preferences. At odds with the original principle of personalized recommendations, which is to give users control of information access, however, the typical and most popular collaborative filtering recommendation technology performs all steps autonomously and allows no room for users to get involved in information personalization of their own. The lacking user involvement causes several problems (illustrated in Section 12.3.1), and many researchers have called attention to solutions to improve the quality of typical recommendation technology and to cope with the relevant problems. Among the several streams of research to develop the solutions, one important evolution is to take advantage of users' self-defined online social networks and proposes to generate recommendations based on users' social linkssocial link-based recommendations. The idea to utilize users' online social links as a foundation of recommendation is undoubtedly spurred by the recent phenomenon of online social networks.Compared with the era when web users stayed in isolation, a number of social media systems have been adapting online sociability, helping users to find people of interest, and as well as encouraging them to socially associate with people of interest. In fact, web users' active and eager participation on social media are intrinsically motivated by not only personal desire for information management and knowledge acquisition but also social desire for engagement and communication with...

show abstract

Section: Regularization Methodsmentioning

confidence: 99%

Section: Input Data Types Of Social Link-based Recommendationsmentioning

confidence: 99%

See 1 more Smart Citation

Recommendations Based on Social Links

Lee

Brusilovsky

2018

Social Information Access

View full text Add to dashboard Cite

show abstract

“…RS is uses by many areas in web such as movies [5], music [6], news [7], tourism [8], social networks [9] and scientific papers [10]. The main approaches of the recommenders includes content based and collaborative filtering where the first one predicts the interests of the user based on the sole user's data whereas the later one predicts the interests of the user based on the similar user's data [11].…”

Section: Related Workmentioning

confidence: 99%

DBT Recommender: Improved Trustworthiness of Ratings through De-Biasing Tendency of Users

Mukkamula¹

2018

IJIES

View full text Add to dashboard Cite

Recent advancements in business strategies marked the significance of e-commerce in marketing any service of the organization. Moreover, users are quiet dependent on the average ratings of the products showcased in the marketing interface in turn these average ratings made remarkable impact on sales phenomena of the product. The average rating of a product is the aggregation of individual users ratings biased with the tendency of the user towards publishing the opinion. The optimistic user tends to give a slight high rating than a neutral judgement and vice versa with a pessimistic user. However, these biased ratings produce an aggregate value that is degraded with its trustworthiness. This paper proposed a novel approach named DBT (De-biased Tendency) Recommender to analyze the bias in product rating which recalculates the average ratings of the products by making user tendencies as part of the process. The solution implemented on a big data environment on demand of high computation complexity involved in the process. Experimental results had shown a significant improvement in the trustworthiness of the product ratings with the proposed approach.

show abstract

“…The purpose of this study was to present a creation of recommender system based on expert and item category because the expert recommendation is more reliable than friends or people [6], and the item category can help improve a performance of recommendation [7]. The researcher also tested an evaluation of performance for error and accuracy in prediction.…”

Section: Introductionmentioning

confidence: 99%

Recommender System Based on Expert and Item Category

Phukseng¹,

Sodsee²

2018

View full text Add to dashboard Cite

Abstract. The objective of this study was to introduce the recommender system based on expert and item category to match the right items to users. In this study, the expert identification was divided into 3 techniques which were 1) the experts from social network technique 2) the experts from the frequency of rating technique and 3) the experts from other user's preferences. To filter the expert users by using the frequency of rating technique and the experts from other user's preferences technique, data about item category is used. For evaluation in this study, the researcher used Epinion for the performance testing to find out errors and accuracies in the prediction process. The results of this study showed that all the presented techniques had mean absolute error score at about 0.15 and 85 percentages of accuracy, especially the expert identification combining with item category, it can reduce 60 percentages of the duration of recommendation creating.

show abstract

Recommender systems based on social networks

Cited by 157 publications

References 7 publications

Recommendations Based on Social Links

Recommendations Based on Social Links

DBT Recommender: Improved Trustworthiness of Ratings through De-Biasing Tendency of Users

Recommender System Based on Expert and Item Category

Contact Info

Product

Resources

About