Multi-Pointer Co-Attention Networks for Recommendation

By Yi Tay, Anh Tuan Luu, Siu Cheung Hui. Knowledge Discovery and Data Mining, 2018

308 citations recorded by the source corpus.

Many recent state-of-the-art recommender systems such as D-ATT, TransNet and DeepCoNN exploit reviews for representation learning. This paper proposes a new neural architecture for recommendation with reviews. Our model operates on a multi-hierarchical paradigm and is based on the intuition that not all reviews are created equal, i.e., only a selected few are important. The importance, however, should be dynamically inferred depending on the current target. To this end, we propose a review-by-review pointer-based learning scheme that extracts important reviews from user and item reviews and subsequently matches them in a word-by-word fashion. This enables not only the most informative reviews to be utilized for prediction but also a deeper word-level interaction. Our pointer-based method operates with a gumbel-softmax based pointer mechanism that enables the incorporation of discrete vectors within differentiable neural architectures. Our pointer mechanism is co-attentive in nature, learning pointers which are co-dependent on user-item relationships. Finally, we propose a multi-pointer learning scheme that learns to combine multiple views of user-item interactions. We demonstrate the effectiveness of our proposed model via extensive experiments on 24 benchmark datasets from Amazon and Yelp. Empirical results show that our approach significantly outperforms existing state-of-the-art models, with up to 19% and 71% relative improvement when compared to TransNet and DeepCoNN respectively. We study the behavior of our multi-pointer learning mechanism, shedding light on 'evidence aggregation' patterns in review-based recommender systems.

Subfields: Artificial Intelligence, Information Systems, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/68eb795891d5c814d2795a82ade1bb8bb211a821

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xu Chen
2. Yongfeng Zhang
3. Naomie Binti Salim
4. Aminu Da’u
5. Yi Tay
6. Anh Tuan Luu
7. Xiangnan He
8. Siu Cheung Hui

In Information Systems

1. Yongfeng Zhang
2. Xu Chen
3. Naomie Binti Salim
4. Aminu Da’u
5. Yi Tay
6. Xiangnan He
7. Zheng Qin
8. Anh Tuan Luu

In Management Science and Operations Research

1. Aminu Da’u
2. Naomie Binti Salim
3. Yi Tay
4. Yongfeng Zhang
5. Siu Cheung Hui
6. Anh Tuan Luu
7. Fei Sun
8. Xing Xie

Citations in the corpus

• A Capsule Network for Recommendation and Explaining What You Like and Dislike (2019, methodology): “Recently, many works are proposed to model contextual information from reviews for better recommendation with deep learning techniques [3, 4, 24, 27, 39].”
• A General Survey on Attention Mechanisms in Deep Learning (2022, background): “example, in [39], a co-attention network is proposed that”
• Graph Neural Networks in Recommender Systems: A Survey (2020, background): “Due to the significance of explainability, many interests have focused on designing explainable recommendation models or conducting post-hoc interpretations [63, 126, 139].”
• Co-Attentive Multi-Task Learning for Explainable Recommendation (2019, methodology): “We calculate an initial review embedding by following [Tay et al., 2018].”
• Personalized Fashion Recommendation with Visual Explanations based on Multimodal Attention Network: Towards Visually Explainable Recommendation (2019, methodology): “On one hand, many meth-ods [5, 9, 31, 36, 40] provide explanations in an "extractive" manner.”
• Progressive Layered Extraction (PLE): A Novel Multi-Task Learning (MTL) Model for Personalized Recommendations (2020, methodology): “To better exploit correlations between tasks, Chen et al. [3] utilize hierarchical multi-pointer co-attention [20] to improve the performance of the recommendation task and explanation task.”
• Deep Learning Based Recommender System (2017, methodology): “Currently, attentional models [127, 147, 179] have more or less eased the noninterpretable concerns of neural models.”
• Dynamic Explainable Recommendation Based on Neural Attentive Models (2019, background): “For providing tailored explanations for different target items, MPCN (Tay, Tuan, and Hui 2018) lever-aged ”co-attention” mechanism to capture the correlations between the users and the items.”
• Recommendation system based on deep learning methods: a systematic review and new directions (2019, methodology): “These show the promising potentials of the deep learning methods for the multitasking RS. Neural Attention Neural attention or attention mechanism is a technique that enables a neural network to focus on a subset of features by selecting a specific input (Tay et al. 2018b).”
• Latent Relational Metric Learning via Memory-based Attention for Collaborative Ranking (2017, background): “ions. There is also an emerging line of work focusing on representation learning using reviews, e.g., Deep Co-operative Networks (DeepCoNN) [43]. A recent work, the Multi-Pointer CoAttention Networks [35] is the state-of-the-art review-based CF model that uses pointer-based attention for representat”
• CoupleNet: Paying Attention to Couples with Coupled Attention for Relationship Recommendation (2018, methodology): “…our approach also follows the neural IR approach which is mainly concerned with modeling document-query pairs (Severyn and Moschitti 2015; Tay et al. 2017; Tay, Tuan, and Hui 2017) or user-item pairs (Zheng, Noroozi, and Yu 2017; Tay, Tuan, and Hui 2018) since we deal with the textual domain.”