Extracting Core Users Based on Features of Users and Their Relationships in Recommender Systems

Extracting Core Users Based on Features of Users and Their Relationships in Recommender Systems

Li Kuang (School of Software, Central South University, Changsha, China), Gaofeng Cao (School of Software, Central South University, Changsha, China) and Liang Chen (School of Data and Computer Science, Sun Yat-Sen University, Guangzhou, China)
Copyright: © 2017 |Pages: 23
DOI: 10.4018/IJWSR.2017040101
OnDemand PDF Download:
List Price: $37.50
10% Discount:-$3.75


As an effective way to solve information overload, recommender system has drawn attention of scholars from various fields. However, existing works mainly focus on improving the accuracy of recommendation by designing new algorithms, while the different importance of individual users has not been well addressed. In this paper, the authors propose new approaches to identifying core users based on trust relationships and interest similarity between users, and the popular degree, trust influence and resource of individual users. First, the trust degree and interest similarity between all user pairs, as well as the three attributes of individuals are calculated. Second, a global core user set is constructed based on three strategies, which are frequency-based, rank-based, and fusion-sorting-based. Finally, the authors compare their proposed methods with other existing methods from accuracy, novelty, long-tail distribution and user degree distribution. Experiments show the effectiveness of the authors' core user extraction methods.
Article Preview

1. Introduction

With the development of Internet technology, various kinds of information online provides colorful and convenient life to people, and they are now used to reading news and updates of friends from Internet, buying books, commodities and various things online, watching movies and listening to music through Internet. However, on the other hand, with the explosion of information, more and more people find it hard to get the information that they really require and are interested in quickly and effectively. In such a background, recommender system is proposed and applied in many online platforms as an effective way to solve the problem of “information overload”. It can provide more personalized services by predicting potential interests according to users’ historical choices. Recommender systems have already been applied widely. For example, Amazon.com uses one’s purchase records to recommend books, Adaptive Info.com uses one’s reading history to recommend news, and the TiVo digital video system recommends TV shows and movies on the basis of users’ viewing patterns and ratings.

The recommender system is very helpful for filtering information, and the core of recommender system is the recommendation algorithm. Recommender system helps users make choices by the way of information filtering, and a successful recommendation to one people may influence subsequent recommendation to other people. The influence is expanded greatly with successive recommendations. From such perspective, it can also explain the evolution of a popular movie.

Recommender systems typically produce a list of recommendations in one of the three ways: content-based (Peng, 2010), collaborative filtering (Chen, 2012; Li, 2010) and context-aware approaches (Adomavicius, 2011; Panniello, 2014). Content-based filtering is also referred to cognitive filtering, which recommends items based on a comparison between the content of the items and the user profiles. Collaborative filtering approaches are based on collecting and analyzing a large amount of information on users’ behaviors, activities or preferences and predicting what users will like based on their similarity to other users. Whereas context-aware approaches consider the influence of context factors, such as natural situation (time, location and etc.) and user’s profile (age, gender, profession and etc.), on users’ demand, preferences, and the selection and definition of neighbors. Based on the basic algorithms, many extensions have been made (Zheng, 2014; Habegger, 2014; Chen, 2010; Zhang, 2014; Xu, 2013; Zhang, 2015).

Complete Article List

Search this Journal:
Volume 20: 1 Issue (2023): Forthcoming, Available for Pre-Order
Volume 19: 4 Issues (2022): 1 Released, 3 Forthcoming
Volume 18: 4 Issues (2021)
Volume 17: 4 Issues (2020)
Volume 16: 4 Issues (2019)
Volume 15: 4 Issues (2018)
Volume 14: 4 Issues (2017)
Volume 13: 4 Issues (2016)
Volume 12: 4 Issues (2015)
Volume 11: 4 Issues (2014)
Volume 10: 4 Issues (2013)
Volume 9: 4 Issues (2012)
Volume 8: 4 Issues (2011)
Volume 7: 4 Issues (2010)
Volume 6: 4 Issues (2009)
Volume 5: 4 Issues (2008)
Volume 4: 4 Issues (2007)
Volume 3: 4 Issues (2006)
Volume 2: 4 Issues (2005)
Volume 1: 4 Issues (2004)
View Complete Journal Contents Listing