Recommender systems based on user reviews: the state of the art

By Li Chen, Guanliang Chen, Feng Wang. User modeling and user-adapted interaction, 2015

386 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Sociology and Political Science

View Paper: https://www.semanticscholar.org/paper/09bf97f066bcfcf2cce2c3e9cc34509f8368765c

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. Yongfeng Zhang
2. Xu Chen
3. Qingyao Ai
4. Zheng Qin
5. Philip S. Yu
6. W. Bruce Croft
7. Lidong Bing
8. B. Liu

In Information Systems

1. Yongfeng Zhang
2. Xu Chen
3. Qingyao Ai
4. Zheng Qin
5. Hongyuan Zha
6. W. Bruce Croft
7. Philip S. Yu
8. B. Liu

In Sociology and Political Science

1. Philip S. Yu
2. W. Bruce Croft
3. Barry Smyth
4. Hongyuan Zha
5. B. Liu
6. Alexander Tuzhilin
7. Yongfeng Zhang
8. Lei Zheng

Citations in the corpus

• Recommendation (2020, neutral): “Recommendation algorithms can be divided into four categories: content-based approaches, rating-based collaborative filtering, preference-based product ranking, and review-based approaches [3].”
• Coevolutionary Recommendation Model: Mutual Learning between Ratings and Reviews (2018, background): “Such techniques can be used to infer user ratings for products and services, and even combined with real ratings and more conventional ratings-based techniques, to generate improved recommendations; for example see [5].”
• Neural Rating Regression with Abstractive Tips Generation for Recommendation (2017, background): “Multi-faceted information can be extracted from reviews and used as user preferences or item features, which otherwise cannot be obtained from the overall ratings [5].”
• Adversarial Distillation for Efficient Recommendation with External Knowledge (2018, background): “Using review information to enhance the recommendation performance is getting increasing attention in both research and industry communities [5].”
• Joint Representation Learning for Top-N Recommendation with Heterogeneous Information Sources (2017, background): “One trend is to augment numerical ratings with textual reviews for recommendation [8], which includes topic modeling [3, 26, 27, 34], sentiment analysis [9, 13, 39, 42–44], and (deep) neural network [2, 36, 41, 45] approaches for review modeling.”
• Aspect Based Recommendations: Recommending Items with the Most Valuable Aspects Based on User Reviews (2017, background): “Much of the work in the area focuses on trying to improve estimations of user ratings of items based on the user reviews and other relevant information [6] and also to explain why some particular recommendations are given to the user based on the review information [25].”
• Why I like it: multi-task learning for recommendation and explanation (2018, background): “To improve the accuracy and explainability of recommender systems, the recommender systems community has recently tried to harness user-generated reviews as a new source of recommendation data [4, 9, 10, 12, 13, 18].”
• Joint Deep Modeling of Users and Items Using Reviews for Recommendation (2017, methodology): “Most of the recommender systems which use reviews in the modeling process employ topic modeling techniques to model users or items [6].”