A social recommendation method based on opinion leaders
- PDF / 1,431,718 Bytes
- 16 Pages / 439.37 x 666.142 pts Page_size
- 34 Downloads / 175 Views
A social recommendation method based on opinion leaders Lijuan Weng 1 & Qishan Zhang 1 Received: 27 November 2019 / Revised: 20 July 2020 / Accepted: 24 September 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
With the rapid development of information technology, social media has been widely used, and Internet information has been exploded, and consumers may experience information overload. Recommender systems using the social recommendation method that integrates social relationship information can provide users with target information that meets their needs. However, most of the existing methods only rely on the user’s ordinary friends to make recommendations, neglecting another influential group, the opinion leaders. In this study, we propose a new social recommendation method based on opinion leaders. The proposed method assumes that the influence of the opinion leader on the user is much greater than that of the user’s ordinary friends. The experimental results on two real datasets show that the proposed method not only has a better recommendation effect than the state-of-the-art recommendation algorithms, but also has a good performance in the cases of cold-start users. Keywords Recommender system . Opinion leaders . Social recommendation . Probabilistic matrix factorization
1 Introduction Recommender systems have received growing research attention from scholars and have been widely used in many industries. The level of recommendation accuracy has been steadily improved in recent years, as problems such as data sparsity, cold start and poor system performance have been overcome or improved. Scholars have started to incorporate the users’ social relationship information (such as label information, concern relationship and trust relationship) into the traditional recommender system [19], the so-called social recommendation methods [22, 29]. Wang et al. [28] consider the information of the user’s friends into the
* Qishan Zhang [email protected]
1
School of Economics and Management, Fuzhou University, Fujian 350108, People’s Republic of China
Multimedia Tools and Applications
recommender system, based on their belief that the user’s preferences would be affected by the taste of their friends. Guo et al. [8] use the trust link between users to improve accuracy. Ma et al. [16] use probabilistic matrix factorization to model the user’s social information. Chen et al. [2] integrate both the target user’s preferences with the impact of neighbors to predict unknown ratings. Most of the social relationship-based methods simply regard users with links or the high similarities among friends, which may lead to biased evaluation of the user’s preferences. Moreover, as the online social networks continue to expand, it is becoming difficult to effectively learn the weight of each link, and the amount of calculation becomes so large that the performance of the recommender system deteriorates. Most of the state-of-the-art research on social recommendation ignores the influence of
Data Loading...
