Attentive Aspect Modeling for Review-Aware Recommendation

Guan, Xinyu; Cheng, Zhiyong; He, Xiangnan; Zhang, Yongfeng; Zhu, Zhibo; Peng, Qinke; Chua, Tat-Seng

doi:10.1145/3309546

Cited by 101 publications

(53 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reviews and ratings, as the most common user-generated information in the e-commerce system, are often combined to make more accurate and interpretable recommendations. The review-based recommendation methods can be generally categorized into sentiment-based, topic-based and deep learning based [7], [10].…”

Section: Related Workmentioning

confidence: 99%

A Review Semantics Based Model for Rating Prediction

2020

View full text Add to dashboard Cite

A review expresses the concerned aspects and corresponding assessments a customer has towards a particular item. Extracting the user's interests and product's features from their aggregated reviews and matching them together to predict the overall rating is a common paradigm in a review-based recommender. However, such a paradigm trains model on the aggregated historical review which grows with time and may have much conflicting semantics, thus the scalability and accuracy may be compromised. In this paper, a novel review semantics based model(RSBM) is proposed to enhance the performance of the review-based recommender. It consists of three parts: the review semantics extractor, the review semantics generator and the rating regressor. Firstly, the review semantics extractor uses a convolutional neural network(CNN) to extract the semantic features of a particular review text. Secondly, the semantics generator uses a memory-network liked structure and attention mechanism to simulate the decision-making process which assesses each concerned aspect of an item to generate the review semantics. In the training phase, the generated semantics is compared with the semantics extracted by the review semantics extractor. In the last, the generated semantic features are fed into the rating regressor to predict the overall rating. Experiments on a series of reality datasets show that the proposed model gains better performances than several state-of-the-art recommendation approaches in terms of accuracy and scalability. INDEX TERMS Recommender system, deep learning.

show abstract

Section: Related Workmentioning

confidence: 99%

A Review Semantics Based Model for Rating Prediction

2020

View full text Add to dashboard Cite

show abstract

“…For instance, using fine-grained aspectlevel sentiment analysis can automatically discover the most valuable aspect to enhance future user experience [35]. The second group, such as LDA [5], [32], AARM [36], FLAME [37], and ASCF [38], builds an internal structure or framework to represent different aspects in a user or item review. Chin et al [4] combined user and item aspect-level representations with aspect importance, and then estimated an overall rating for any user-item pair.…”

Section: B Aspect-based Recommendationmentioning

confidence: 99%

Aspect-Based Fashion Recommendation With Attention Mechanism

2020

IEEE Access

View full text Add to dashboard Cite

With the rapid growth of fashion e-commerce, fashion recommendation has become a main digital marketing tool that is built on customer reviews and ratings. Online review is a powerful source for understanding users' shopping experiences, preferences and feedbacks on product/item performances, and thus is useful for enhancing personalized recommendations for future purchases. However, most extant fashion recommendation methods lack effective frameworks to integrate local and global aspect representations extracted from customers' ratings and reviews. In this paper, we proposed an aspect-based fashion recommendation model with attention mechanism (AFRAM) to predict customer ratings based on online reviews of fashion products. This model can extract latent aspect features about users and items separately through two parallel paths of convolutional neural networks (CNN), long short-term memory networks (LSTM), and attention mechanisms. One path processes user reviews and the other copes with item reviews. On each path, CNN and LAST are both coupled with an attention mechanism to capture local aspect features and global aspect features respectively, which are combined through a mutual operation module. The extracted features from the two paths are further merged through another mutual operation to predict users' ratings. The mutual operations on both paths can enhance the generalization of the AFRAM model. Realworld customer reviews and ratings collected from two renowned business websites were used to train and test AFRAM. The experiment results demonstrate that AFRAM is more effective in customer rating predictions, as compared to several state-of-the-art fashion recommenders.

show abstract

“…Toward this end, Bahdanau et al [2] introduced the general attention mechanism working on identifying the important words from auxiliary textual information to provide more precise representations for data. Since then, many derivatives of the attention mechanism have been proposed to solve various tasks from the natural language processing domain [19,21,27,68,69,85,87] and the computer vision domain [61-63, 79, 90]. To be specific, in the natural language processing domain, Yin et al [87] presented three attention schemes to incorporate the mutual influence between sentences into CNNs to learn the sentence representations.…”

Section: Representation Learningmentioning

confidence: 99%

Fine-Grained Privacy Detection with Graph-Regularized Hierarchical Attentive Representation Learning

Chen

Song

Ren

et al. 2020

ACM Trans. Inf. Syst.

Self Cite

View full text Add to dashboard Cite

Due to the complex and dynamic environment of social media, user generated contents (UGCs) may inadvertently leak users' personal aspects, such as the personal attributes, relationships and even the health condition, and thus place users at high privacy risks. Limited research efforts, thus far, have been dedicated to the privacy detection from users' unstructured data (i.e., UGCs). Moreover, existing efforts mainly focus on applying conventional machine learning techniques directly to traditional hand-crafted privacy-oriented features, ignoring the powerful representing capability of the advanced neural networks. In light of this, in this article, we present a fine-grained privacy detection network (GrHA) equipped with graph-regularized hierarchical attentive representation learning. In particular, the proposed GrHA explores the semantic correlations among personal aspects with graph convolutional networks to enhance the regularization for the UGC representation learning, and, hence, fulfil effective fine-grained privacy detection. Extensive experiments on a real-world dataset demonstrate the superiority of the proposed model over state-of-the-art competitors in terms of eight standard metrics. As a byproduct, we have released the codes and involved parameters to facilitate the research community. CCS Concepts: • Information systems → Retrieval tasks and goals; • Security and privacy → Privacy protections;

show abstract

Attentive Aspect Modeling for Review-Aware Recommendation

Cited by 101 publications

References 50 publications

A Review Semantics Based Model for Rating Prediction

A Review Semantics Based Model for Rating Prediction

Aspect-Based Fashion Recommendation With Attention Mechanism

Fine-Grained Privacy Detection with Graph-Regularized Hierarchical Attentive Representation Learning

Contact Info

Product

Resources

About