Abstract
When information from traditional recommender systems is augmented with information about user relationships that social networks store, more successful recommendations can be produced. However, this information regarding user relationships may not always be available, since some users may not consent to the use of their social network information for recommendations or may not have social network accounts at all. Moreover, the rating data (categories and characteristics of products) may be unavailable for a recommender system. In this paper, we present an algorithm that can be applied in any social network-aware recommender system that utilizes the users' ratings on items and users' social relations. The proposed algorithm addresses the issues of limited social network information or limited collaborative filtering information for some users by adapting its behavior, taking into account the density and utility of each user's social network and collaborative filtering neighborhoods. Through this adaptation, the proposed algorithm achieves considerable improvement in rating prediction accuracy. Furthermore, the proposed algorithm can be easily implemented in recommender systems.
Highlights
Recommender systems (RSs) continuously augment their information repositories with data from diverse sources, ranging from smartphones and wearable devices to websites and social networks (SNs), to promote the formulation of successful personalized recommendations in a wide range of domains, from consumer products, such as books, office supplies and CDs, to travel and leisure as well as from restaurants and movies to smartphone apps [1]
In the context of Collaborative filtering (CF), personalization is achieved by formulating recommendations for any user U based on the opinions of other users that have rated items similar to U : this approach is underpinned by the CF fundamental assumption that if users U1 and U2 have similar behavior on some items, they will act on others [2]
The algorithm formulates two partial prediction scores, from the CF and the SN neighborhood, and combines these two partial predictions using a weighted average metascore combination technique, where the weights are personalized for each individual user
Summary
Recommender systems (RSs) continuously augment their information repositories with data from diverse sources, ranging from smartphones and wearable devices to websites and social networks (SNs), to promote the formulation of successful personalized recommendations in a wide range of domains, from consumer products, such as books, office supplies and CDs, to travel and leisure as well as from restaurants and movies to smartphone apps [1]. D. Margaris et al.: Adaptive Social Network-Aware Collaborative Filtering Algorithm for Improved Rating Prediction Accuracy general acceptance, and social influence [4] to supplement the traditional CF data (such as item static characteristics and user ratings). Margaris et al.: Adaptive Social Network-Aware Collaborative Filtering Algorithm for Improved Rating Prediction Accuracy general acceptance, and social influence [4] to supplement the traditional CF data (such as item static characteristics and user ratings) By taking this information into consideration in the recommendation process, the SN RSs manage to achieve better user-targeted recommendations. The proposed algorithm addresses the issues of limited SN information or limited CF information for some users by adapting its behavior, taking into account the density and utility of each user’s SN and CF neighborhoods Through this adaptation, which optimizes the system for individual users [6], the proposed algorithm achieves considerable improvement in rating prediction accuracy.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
