Published on Wed Aug 18 2021

SIFN: A Sentiment-aware Interactive Fusion Network for Review-based Item Recommendation

Kai Zhang, Hao Qian, Qi Liu, Zhiqiang Zhang, Jun Zhou, Jianhui Ma, Enhong Chen

We propose a Sentiment-aware Interactive Fusion Network (SIFN) for review-based item recommendation. We first encode user/item reviews via BERT and propose a light-weighted sentiment learner to extract semantic features of each review. Then, we propose a sentiment prediction

1
0
0
Abstract

Recent studies in recommender systems have managed to achieve significantly improved performance by leveraging reviews for rating prediction. However, despite being extensively studied, these methods still suffer from some limitations. First, previous studies either encode the document or extract latent sentiment via neural networks, which are difficult to interpret the sentiment of reviewers intuitively. Second, they neglect the personalized interaction of reviews with user/item, i.e., each review has different contributions when modeling the sentiment preference of user/item. To remedy these issues, we propose a Sentiment-aware Interactive Fusion Network (SIFN) for review-based item recommendation. Specifically, we first encode user/item reviews via BERT and propose a light-weighted sentiment learner to extract semantic features of each review. Then, we propose a sentiment prediction task that guides the sentiment learner to extract sentiment-aware features via explicit sentiment labels. Finally, we design a rating prediction task that contains a rating learner with an interactive and fusion module to fuse the identity (i.e., user and item ID) and each review representation so that various interactive features can synergistically influence the final rating score. Experimental results on five real-world datasets demonstrate that the proposed model is superior to state-of-the-art models.

Sun Jan 28 2018
Artificial Intelligence
Multi-Pointer Co-Attention Networks for Recommendation
Many recent state-of-the-art recommender systems exploit reviews for representation learning. We propose a review-by-review pointer-based learning scheme that extracts important reviews. This enables not only the most informative reviews to be utilized for prediction but also a deeper word-level interaction.
0
0
0
Thu Oct 15 2020
Artificial Intelligence
Hierarchical Text Interaction for Rating Prediction
Traditional recommender systems encounter challenges such as data sparsity and unexplained recommendation. We propose a novel Hierarchical Text Interaction model(HTI) for rating prediction. In HTI, we model low-level word semantics and high-level review representations hierarchically.
0
0
0
Tue Jan 17 2017
Machine Learning
Joint Deep Modeling of Users and Items Using Reviews for Recommendation
A large amount of information exists in reviews written by users. This source of information has been ignored by most of the current recommender systems. We present a deep model to learn item properties and user behaviors jointly from review text.
0
0
0
Wed Jan 24 2018
Machine Learning
Improving Review Representations with User Attention and Product Attention for Sentiment Classification
Neural network methods have achieved great success in reviews sentiment classification. Some words or sentences show strong user's preference, while others indicate product's characteristic. It is not reasonable to encode user and product information together into one representation.
0
0
0
Mon Nov 02 2020
Machine Learning
Hierarchical Bi-Directional Self-Attention Networks for Paper Review Rating Recommendation
Review rating prediction of text reviews is a rapidly growing technology with a wide range of applications in natural language processing. Most existing methods either use hand-crafted features or learn features using deep learning with simple text corpus as input. We propose a Hierarchical bi-directional self-attention
0
0
0
Mon Feb 06 2017
NLP
Opinion Recommendation using Neural Memory Model
opinion recommendation is a novel task of jointly predicting a customer review with a rating score that a certain user would give to a certain product or service. We use a single neural network to model users and products, capturing their correlation and generating customised product representations using a deep memory network. Results show
0
0
0