Paper information
- title: Deep Modeling of Social Relations for Recommendation
- authors: Wenqi Fan, Qing Li, Min Cheng
- venue: AAAI 18
- pdf link: https://paperpile.com/shared/YBHUxN
- github:
- abstract:
Social-based recommender systems have been recently proposed by incorporating social relations of users to alleviate sparsity issue of user-to-item rating data and to improve recommendation performance. Many of these social-based recommender systems linearly combine the multiplication of social features between users. However, these methods lack the ability to capture complex and intrinsic non-linear features from social relations. In this paper, we present a deep neural network based model to learn non-linear features of each user from social relations, and to integrate into probabilistic matrix factorization for rating prediction problem. Experiments demonstrate the advantages of the proposed method over stateof-the-art social-based recommender systems.
Summary:
problems to address
Many of the social-based recommender systems linearly combine the multiplication of social features between users. However, these methods lack the ability to capture complex and intrinsic non-linear features from social relations.
key ideas
present a deep neural network based model to learn non-linear features of each user from social relations, and to integrate into probabilistic matrix factorization for rating prediction problem.
-
About probabilistic matrix factorization [1]
Problem to solve: Many existing approaches to collaborative filtering can neither handle very large datasets nor easily deal with users who have very few ratings.
The Probabilistic Matrix Factorization (PMF) model which scales linearly with the number of observations and, more importantly, performs well on the large, sparse, and very imbalanced Netflix dataset.
-
This paper
By replacing the user latent models of the probabilistic matrix factorization(PMF) (Salakhutdinov and Mnih 2007) with DNN Net(u) user latent feature vector u in our proposed model is approximated by social latent vector generated from DNN as follows
More detail pls see https://docs.google.com/document/d/18lmzg3q97p2ks8OX4GOBAQUU6txkaB9y59KdO-hJImI/edit?usp=sharing
Questions about the paper?
What do you like?
Combine DNN and the probabilistic model.
Add social relationships to model users.
What you don't like?
The comparison only includes some papers related to social relations and some early work, without some comparison with other SOTA methods, and the performance is also not very good.
How to improve?
Graph neural network for social relationships [2]
Any new ideas?
Reproducing results (if any)
Reference:
[1]Mnih, Andriy, and Russ R. Salakhutdinov. 2008. “Probabilistic Matrix Factorization.” In Advances in Neural Information Processing Systems
[2]Wenqi Fan, Yao Ma, Qing Li, Yuan He, Eric Zhao, Jiliang Tang, Dawei Yin, Graph Neural Networks for Social Recommendation,WWW '19: The World Wide Web Conference
Paper information
Social-based recommender systems have been recently proposed by incorporating social relations of users to alleviate sparsity issue of user-to-item rating data and to improve recommendation performance. Many of these social-based recommender systems linearly combine the multiplication of social features between users. However, these methods lack the ability to capture complex and intrinsic non-linear features from social relations. In this paper, we present a deep neural network based model to learn non-linear features of each user from social relations, and to integrate into probabilistic matrix factorization for rating prediction problem. Experiments demonstrate the advantages of the proposed method over stateof-the-art social-based recommender systems.
Summary:
problems to address
Many of the social-based recommender systems linearly combine the multiplication of social features between users. However, these methods lack the ability to capture complex and intrinsic non-linear features from social relations.
key ideas
present a deep neural network based model to learn non-linear features of each user from social relations, and to integrate into probabilistic matrix factorization for rating prediction problem.
About probabilistic matrix factorization [1]
Problem to solve: Many existing approaches to collaborative filtering can neither handle very large datasets nor easily deal with users who have very few ratings.
The Probabilistic Matrix Factorization (PMF) model which scales linearly with the number of observations and, more importantly, performs well on the large, sparse, and very imbalanced Netflix dataset.
This paper
By replacing the user latent models of the probabilistic matrix factorization(PMF) (Salakhutdinov and Mnih 2007) with DNN Net(u) user latent feature vector u in our proposed model is approximated by social latent vector generated from DNN as follows
More detail pls see https://docs.google.com/document/d/18lmzg3q97p2ks8OX4GOBAQUU6txkaB9y59KdO-hJImI/edit?usp=sharing
Questions about the paper?
What do you like?
Combine DNN and the probabilistic model.
Add social relationships to model users.
What you don't like?
The comparison only includes some papers related to social relations and some early work, without some comparison with other SOTA methods, and the performance is also not very good.
How to improve?
Graph neural network for social relationships [2]
Any new ideas?
Reproducing results (if any)
Reference:
[1]Mnih, Andriy, and Russ R. Salakhutdinov. 2008. “Probabilistic Matrix Factorization.” In Advances in Neural Information Processing Systems
[2]Wenqi Fan, Yao Ma, Qing Li, Yuan He, Eric Zhao, Jiliang Tang, Dawei Yin, Graph Neural Networks for Social Recommendation,WWW '19: The World Wide Web Conference