A hybrid neural network approach to combine textual information and rating information for item recommendation
- PDF / 2,912,142 Bytes
- 26 Pages / 439.37 x 666.142 pts Page_size
- 63 Downloads / 225 Views
A hybrid neural network approach to combine textual information and rating information for item recommendation Donghua Liu1 · Jing Li1 · Bo Du1 · Jun Chang1 · Rong Gao2 · Yujia Wu1 Received: 24 April 2018 / Revised: 30 October 2020 / Accepted: 1 November 2020 © Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract Collaborative filtering (CF) is a common method used by many recommender systems. Traditional CF algorithms exploit users’ ratings as the sole information source to learn user preferences. However, ratings usually sparse cause a serious impact on the recommendation results. Most existing CF algorithms use ratings and textual information to alleviate the sparsity of data and then utilize matrix factorization to achieve the latent feature interactions for rating prediction. Nevertheless, the following shortcomings remain in these studies: (1) The word orders and surrounding words of the textual information are ignored. (2) The nonlinearity of feature interactions is seldom exploited. Therefore, we propose a novel hybrid neural network to combine textual information and rating (NCTR) information for item recommendation. The proposed NCTR model is built upon a hybrid neural network framework with fine-grained modeling of latent representation and nonlinearity feature interactions for rating prediction. Specifically, convolution neural network is applied to extract effectively contextual features from textual information. Meanwhile, a fusion layer is exploited to combine features, and the multilayer perceptions are used to model the nonlinear interactions between the merged item latent features and user latent features. Experimental results over five real-world datasets show that NCTR significantly outperforms several state-of-theart recommendation methods. Source codes are available in https://github.com/luojia527/ NCTR_master. Keywords Recommender systems · Collaborative filtering · Textual information · Matrix factorization · Neural network
1 Introduction With the rapid development of information technology, massive amounts of information are generated by people and machines. Due to this explosive growth, information overload has emerged. As a result, individuals have to face excessive information, which makes it difficult to find the information that they are interested in. In this situation, recommender systems have emerged as an effective mechanism for providing personalized recommendation services, which can effectively alleviate the information overload problem. Recommender systems
Extended author information available on the last page of the article
123
D. Liu et al.
have attracted much attention in both academia and industry and have been successfully applied in many fields, such as music [27,40], e-commerce [32,51], geographic location [10,24], and others [7,46]. Recommender systems have become an indispensable part of various application services. According to the types of input data, existing recommendation algorithms broadly divide into three categories [3]: content-based algo
Data Loading...
