Trust-embedded collaborative deep generative model for social recommendation
- PDF / 1,972,935 Bytes
- 29 Pages / 439.37 x 666.142 pts Page_size
- 37 Downloads / 182 Views
Trust‑embedded collaborative deep generative model for social recommendation Xiaoyi Deng1 · Yenchun Jim Wu2 · Fuzhen Zhuang3
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Social networks can provide massive amounts of information for communication among users and communities. The trust relationships in social networks can be utilized to reveal user preferences for improving the quality of social recommendation, which aims to mitigate information overload and provide users with the most attractive and relevant items or services. However, the data sparsity and cold-start issue degrade recommendation performance significantly. To address these issues, a novel trust-embedded collaborative deep generative model (TCDG) is proposed for exploiting multisource information (content, rating and trust) to predict ratings. TCDG employs deep generative model to unsupervisedly learn deep latent representations for item content through an inference network in latent space instead of observation space. Meanwhile, TCDG adopts probabilistic matrix factorization to map users into low-dimensional latent feature spaces by trust relationships, which can reflect users’ mutual influence on the formation of users’ opinions more accurately and learn better implicit relationships between items and users from content, rating and trust. In addition, an approach with an annealing parameter to calculate the maximum a posteriori estimates is proposed to learn model parameters. Experiments using four real-world datasets are conducted to evaluate the prediction and top-ranking performance of our model. The results indicate that TDCG has better accuracy and robustness than other methods for making recommendations. Keywords Recommender system · Deep generative model · Collaborative topic regression · Trust matrix factorization · Deep learning
* Yenchun Jim Wu [email protected] Extended author information available on the last page of the article
13
Vol.:(0123456789)
X. Deng et al.
1 Introduction Social networks, such as Facebook, Twitter and Weibo, have become popular platforms that enable communication among users and make it easy for users to generate and share contents among each other [1]. Social networks have significantly potential impacts from different perspectives, such as promoting sales for business profit, improving intellectual capital on firm performance and facilitating individual learning [2, 3]. However, massive amounts of information in social networks can lead to information overload issues, which has become the obstacle for users locating useful contents [4]. The need to solve the problem of information overload has caused the prevalence of recommender systems, which are the most successful and effective application of recommendation. Recommender systems receive users’ information from their regarding items of interest and recommend items which may meet their specific needs. The core of recommender systems typically depends upon a most prevalent and widely adopted methods, collaborat
Data Loading...
