Leveraging multiviews of trust and similarity to enhance clustering-based recommender systems

Guo, Guibing; Zhang, Jie; Yorke-Smith, Neil

doi:10.1016/j.knosys.2014.10.016

Cited by 132 publications

(47 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These approaches though operate strictly on the user-item ratings. More recently, approaches that take into account additional evidence have been developed, including a multiview clustering method that iteratively clusters users on the basis of both rating patterns and social trust relationships [9], as well as a co-clustering technique that exploits item interactions and relationships, together with user community information [22].…”

Section: Related Workmentioning

confidence: 99%

A Hybrid Recommendation System Based on Density-Based Clustering

Tsikrika¹,

Symeonidis²,

Gialampoukidis³

et al. 2018

Internet Science

View full text Add to dashboard Cite

Abstract. Collaboratve filtering recommenders leverage past user-item ratings in order to predict ratings for new items. One of the most critical steps in such methods corresponds to the formation of the neighbourhood that contains the most similar users or items, so that the ratings associated with them can be employed for predicting new ratings. This work proposes to perform the combination of content-based and ratings-based evidence during the neighbourhood formation step and thus identify the most similar neighbours in a hybrid manner. To this end, DBSCAN, a density-based clustering approach, is applied for identifying the most similar users or items by considering the ratings-based and the contentbased similarities, both individually and in combination. The resulting hybrid cluster-based CF recommendation scheme is then evaluated on the latest small MovieLens100k dataset and the experimental results indicate the potential of the proposed approach.

show abstract

Section: Related Workmentioning

confidence: 99%

A Hybrid Recommendation System Based on Density-Based Clustering

Tsikrika¹,

Symeonidis²,

Gialampoukidis³

et al. 2018

Internet Science

View full text Add to dashboard Cite

show abstract

“…Then we combine it into the collaborative recommendation model based on matrix decomposition by factoring such relationship in order to optimize target function by means of random gradient descent algorithm [7], [13], [15], [16]. It is explained as follows 1) Measurement of Trust: Trust-ability is the degree of trust between users.…”

Section: International Journal For Research In Applied Science and Engimentioning

confidence: 99%

Trust Based Novel Recommendation Regularized with Item Ratings

Priyadharshini¹

2017

IJRASET

View full text Add to dashboard Cite

Recommendation is an opinion given by an analyst to his/her client whether the given stock is worth buying or a particular place is worth visiting or not. They use various projections as a basis for issuing recommendations. Item rating is a group of classifications designed to extract information about a quantitative or qualitative attribute. Here we use a scale to reflect the quality of product where user selects the number which is taken into consideration. In order to enhance the novel recommendation model, we propose a trust based recommendation model with item rating where data sparsity and cold start problem are rectified.We make use of personalized social networking to connect people in a commodity so that people can get to know about a product or place in detail by the information shared about it and the user can sort out things according to their needs and specification. Keywords: TrustSVD, Itemratings, Filtration,Recommendation,Privacy preservance. 1.INTRODUCTION Recommender systems have been widely used to provide users with high-quality personalized recommendations from a large volume of choices. In order to reduce the data sparsity and cold start problems and their degradation of recommendation performance we use Trust SVD integrates multiple information sources into the recommendation model. An analysis of trust based social network data from four real-world data sets suggests that not only the explicit but also the implicit influence of both ratings and trust should be taken into consideration in a recommendation model. Trust SVD therefore builds on top of a state-of-the-art recommendation algorithm, SVD++ (which uses the explicit and implicit influence of rated items), by further joining both the explicit and implicit influence of trusted[1] and trusting users on the prediction of items for an active user. The proposed technique is to extend SVD++ with social trust information. Experimental results on the four data sets exhibit that Trust SVD achieves better accuracy than other counter parts of recommendation techniques. A. Role of Apache ServerApache Server is one of the popular open source software organizations. It is free and commercial friendly -no licensing fees or costs. It runs pretty much on any OS (Linux, Windows and MacOS). It is maintained regularly with the norms. It is one of the most feature rich web servers available. By nomenclature "Apache" is Apache HTTP Server (sometimes also called Apache httpd -after the name of the process). The other "Apache" server is Apache Tomcat. B. Role of MySQLMySQL is a free, open-source database management system (DBMS for short). A DBMS is a system that manages databases and connects them to server. For example, a MySQL database can be used to run a website, to run the database of an ERP or any other software. II.RELATED WORK In existing trust based social network provides an alternative view of user preferences rather than item ratings. Find that trust social networks are small-world networks where two random users are socially connecte...

show abstract

“…The approach combines values of users' similarities according to their ratings and similarity values obtained from their genre interests. Guo, Zhang, and Yorke-Smith (2015) perform additional inquiries by proposing a new information source called prior ratings, and design a user study to validate the conceptual model of prior ratings. One of major concerns is how the system should provide recommendations when there are no prior ratings.…”

Section: Related Workmentioning

confidence: 99%