Community detection in social recommender systems: a survey
- PDF / 1,231,028 Bytes
- 21 Pages / 595.224 x 790.955 pts Page_size
- 40 Downloads / 255 Views
Community detection in social recommender systems: a survey Fabio Gasparetti1
· Giuseppe Sansonetti1 · Alessandro Micarelli1
Accepted: 18 September 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Information extracted from social network services promise to improve the accuracy of recommender systems in various domains. Against this background, community detection techniques help us understand more of users’ collective behavior by clustering similar users w.r.t. their interests, preferences and activities. The purpose of this paper is to bring the novice or practitioner quickly up to date with the main outcomes and research directions in the field of social recommendation based on community detection. The research synthesis consists of a narrative review which identifies what has been written on the topic of community-based recommender system. The comprehensive search of relevant literature aims at synthesizing prior study findings by identifying approaches that follow similar paradigms and techniques. The paper is of value to those involved with recommender systems and social media. Keywords Community detection · Recommender systems · Social network services
1 Introduction Popular online social networks, such as Facebook or Twitter, consist of nodes representing people or other entities embedded in a social context, and edges referring to interactions, collaborations, or some other form of linkage between users and entities. The power and appeal of social networking are such that the number of active users worldwide is continuously increasing [116]. Meanwhile Recommender systems (RS) have been gathering increasingly crucial role and influence on our daily life. They are implemented in many web applications for guiding user choices, such as online retailers for recommending items to buy or news articles to read. RS are mostly based on representations of user preferences and interests. Social information, such as contacts and interactions between peers provide a large breadth of signals useful in identifying these aspects of the users. Indeed, research activity on Social recommender systems (SRS) focuses on tapping social information in the ranking process [50], in order to provide more accurate and unexpected recommendations,
Fabio Gasparetti
[email protected] 1
Roma Tre University, Rome, Italy
potentially more efficiently, with respect to more traditional approaches. For instance, Collaborative filtering (CF) based recommenders are often based on user-user similarity measures which can be improved by analyzing the social relationships in the online networks (see Section 4.4). The common rationale is that the likelihood of accepting people’s advice is affected by tie strength, trustworthiness, and homophily towards them [4]. Hence, trusted recommendations from people with similar tastes are considered more than those from strangers and vendors. Indeed, representations of user preferences and interests may be enhanced by analyzing the characteristics of both the use
Data Loading...
