Representation learning via Dual-Autoencoder for recommendation
Introduction
In order to tackle the information overload problem, recommender systems are proposed to help users to find objects of interest by utilizing the user–item interaction information and/or content information associated with users and items. Recommender systems have attracted much attention from multiple disciplines, and many techniques have been proposed to build recommender systems (Adomavicius and Tuzhilin, 2005, Bell, 2011). It is also widely used in many E-commerce companies, such as for product sale on Amazon and movie rentals from Netflix (Srebro & Jaakkola, 2003).
Traditional recommender systems normally only utilize user–item rating feedback information for recommendations. Moreover, the matrix factorization technique is widely used in recommender systems, which factorizes a user–item rating matrix into two low rank user-specific and item-specific matrices, and then utilizes the factorized matrices to make further predictions (Koren et al., 2009, Srebro and Jaakkola, 2003). In order to comprehensively utilize rich information in recommender system, there is a surge of hybrid recommendation, such as social recommendation (Ma et al., 2008, Ma et al., 2011), location based recommendation (Lian et al., 2014, Liu et al., 2013), and heterogeneous network based recommendation (Shi et al., 2015, Yu et al., 2014). Most of these methods are based on a matrix factorization framework, in which the latent factors of users and items are usually obtained by directly factorizing the user–item rating matrix and additional information are usually used as a regularization constraint. Although these methods pay much attention to exploit additional information, we wonder they might not make full use of the user–item rating information. In other words, we might be able to obtain better latent factors of users and items through extensively exploiting rating information.
On the other hand, deep learning has shown its power in learning latent feature representation in many domains, such as image/video processing (Alex et al., 2009) and text data (Socher, Huang, Pennington, Ng, & Manning, 2011). Can we use deep learning techniques to learn latent representations for recommendation? Some researchers have pursued this goal. For example, the latent factor of music is extracted from audio signals with a deep convolutional neural network (Oord, Dieleman, & Schrauwen, 2013) and the tabular data is modeled through the adaption of Restricted Boltzmann Machines (Salakhutdinov, Mnih, & Hinton, 2007). Recently, Wang, Wang, and Yeung (2014) designed a collaborative deep learning method to utilize item content information. In essence, these methods utilize the powerful representation learning of deep learning to analyze the additional information (e.g., audio signal and text context), not rating information, for better recommendations. They did not directly learn the latent factors of users and items with deep learning. Moreover, the additional information sometimes is not easy to acquire and very sparse.
To the best of our knowledge, there has been little effort focused on employing deep learning for recommendations only on user–item rating information. Motivated by the success of the latent feature representation of deep learning on image and text data, we design a novel Recommendation framework via Dual-Autoencoders (ReDa), which is illustrated in Fig. 1. In this figure, ReDa simultaneously learns the new hidden representations of users and items using autoencoders, and minimizes the deviations of training data by the learnt representations of users and items. Moreover, a gradient descent method is derived to learn the hidden representations. Experiments on four real-world data sets demonstrate the effectiveness of our proposed model.
The remainder of this paper is organized as follows. We introduce the notations and preliminary knowledge in Section 2, and then propose the representation learning framework based on autoencoders for recommendation in Section 3. Extensive experiments conducted on several data sets are shown in Section 4, followed by the related work in Section 5 and conclusions in Section 6.
Section snippets
Notations and preliminaries
In this section, we first introduce some frequently used notations as presented in Table 1, and some preliminaries which will be used in our proposed framework.
Representation learning via Dual-Autoencoders
In this section, we first formulate the representation learning framework via dual-autoencoders for recommendation, and then derive the model solution using the gradient decent method.
Experiments
In this section, we conduct experiments on four real-world data sets to systemically evaluate the effectiveness of our proposed framework for recommendation.
Related work
Recent years have witnessed a boom of research work in recommendation systems. A number of techniques are employed for recommendations and many sources of data are fused to improve recommendation performances. Traditional recommender systems normally only utilize user–item rating feedback information for recommendation. Collaborative filtering is one of the most popular techniques, whose basic idea is to find similar objects for recommendation through interactive records. Recently, matrix
Conclusion and future work
To make full use of the user–item rating information and learn better latent representations, different from previous matrix factorization methods we aim to propose a new representation learning model based on autoencoders for recommendation in this paper. In our proposed framework, we simultaneously learn the latent factors of users and items, and minimize the derivations of training data using the learnt latent factors. Experiments on four data sets validate the superiority of our proposed
Acknowledgments
This work is supported by the National Natural Science Foundation of China (Nos. 61473273, 91546122, 61573335, 61602438), Guangdong provincial science and technology plan projects (No. 2015 B010109005), the Youth Innovation Promotion Association CAS 2017146 and 2015 Microsoft Research Asia Collaborative Research Program.
References (24)
- et al.
Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions
IEEE Transactions on Knowledge and Data Engineering
(2005) - et al.
A novel connectionist system for unconstrained handwriting recognition
IEEE Transactions on Pattern Analysis and Machine Intelligence
(2009) Learning deep architectures for ai
Foundations and Trends in Machine Learning
(2009)Factorization meets the neighborhood: A multifaceted collaborative filtering model
- et al.
- et al.
Matrix factorization techniques for recommender systems
Computer
(2009) - Lee, J., Bengio, S., Kim, S., Lebanon, G., & Singer, Y. (2014). Local collaborative ranking for recommendation. In...
- et al.
Algorithms for non-negative matrix factorization
- Lian, Defu, Zhao, Cong, Xie, Xing, Sun, Guangzhong, Chen, Enhong, & Rui, Yong (2014). Geomf: joint geographical...
- et al.
A probabilistic rating auto-encoder for personalized recommender systems
Learning to recommend with social trust ensemble
Cited by (113)
GraphRec-based Korean expert recommendation using author contribution index and the paper abstracts in marine
2024, Engineering Applications of Artificial IntelligenceAn effective knowledge graph entity alignment model based on multiple information
2023, Neural NetworksNEFM: Neural embedding based factorization machines for user response prediction
2023, Expert Systems with ApplicationsCitation Excerpt :In this paper, to address these shortcomings, we propose a neural embedding factorization machine (NEFM) model, which enhances FM models in modeling higher-order and non-linear feature interactions. As been reported in Bengio (2009) and Zhuang et al. (2017), an auto-encoder is powerful to capture feature representation for input data. Being inspired by Item2Vec (Barkan & Koenigstein, 2016), which draws on the idea behind neural embedding (Mikolov, Chen, Corrado and Dean, 2013; Mikolov, Sutskever, Chen, Corrado and Dean, 2013), and learns item representations via a probabilistic auto-encoder, NEFM can effectively initialize the embedding layer of FM models.
ExpGCN: Review-aware Graph Convolution Network for explainable recommendation
2023, Neural NetworksHandling data sparsity via item metadata embedding into deep collaborative recommender system
2022, Journal of King Saud University - Computer and Information Sciences