Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Review-based Recommender Systems: A Survey of Approaches, Challenges and Future Perspectives

Emrul Hasan, Mizanur Rahman, Chen Ding, Jimmy Xiangji Huang, Shaina Raza

TL;DR

This survey addresses a gap in the literature by focusing on review-based recommender systems (RBRS), detailing how textual reviews and aspect information augment traditional rating-based approaches. It introduces two complementary classification schemes for RBRS (review utilization strategies and methodological families) and provides a comprehensive taxonomy spanning probabilistic, deep learning, and miscellaneous methods. The paper synthesizes state-of-the-art approaches, datasets, evaluation metrics, and real-world deployments, highlighting performance patterns (e.g., RMCL and MAGCL) and domain-dependent trade-offs. It further identifies major challenges—representation learning, integration strategies, scalability, and ethics—and outlines future directions such as multimodal data fusion, multi-criteria ratings, real-time adaptation, and the role of Large Language Models in RBRS. The work aims to guide researchers and practitioners toward robust, interpretable, and responsible RBRS with broad impact across e-commerce, hospitality, and media domains.

Abstract

Recommender systems play a pivotal role in helping users navigate an overwhelming selection of products and services. On online platforms, users have the opportunity to share feedback in various modes, including numerical ratings, textual reviews, and likes/dislikes. Traditional recommendation systems rely on users explicit ratings or implicit interactions (e.g. likes, clicks, shares, saves) to learn user preferences and item characteristics. Beyond these numerical ratings, textual reviews provide insights into users fine-grained preferences and item features. Analyzing these reviews is crucial for enhancing the performance and interpretability of personalized recommendation results. In recent years, review-based recommender systems have emerged as a significant sub-field in this domain. In this paper, we provide a comprehensive overview of the developments in review-based recommender systems over recent years, highlighting the importance of reviews in recommender systems, as well as the challenges associated with extracting features from reviews and integrating them into ratings. Specifically, we present a categorization of these systems and summarize the state-of-the-art methods, analyzing their unique features, effectiveness, and limitations. Finally, we propose potential directions for future research, including the integration of multimodal data, multi-criteria rating information, and ethical considerations.

Review-based Recommender Systems: A Survey of Approaches, Challenges and Future Perspectives

TL;DR

This survey addresses a gap in the literature by focusing on review-based recommender systems (RBRS), detailing how textual reviews and aspect information augment traditional rating-based approaches. It introduces two complementary classification schemes for RBRS (review utilization strategies and methodological families) and provides a comprehensive taxonomy spanning probabilistic, deep learning, and miscellaneous methods. The paper synthesizes state-of-the-art approaches, datasets, evaluation metrics, and real-world deployments, highlighting performance patterns (e.g., RMCL and MAGCL) and domain-dependent trade-offs. It further identifies major challenges—representation learning, integration strategies, scalability, and ethics—and outlines future directions such as multimodal data fusion, multi-criteria ratings, real-time adaptation, and the role of Large Language Models in RBRS. The work aims to guide researchers and practitioners toward robust, interpretable, and responsible RBRS with broad impact across e-commerce, hospitality, and media domains.

Abstract

Recommender systems play a pivotal role in helping users navigate an overwhelming selection of products and services. On online platforms, users have the opportunity to share feedback in various modes, including numerical ratings, textual reviews, and likes/dislikes. Traditional recommendation systems rely on users explicit ratings or implicit interactions (e.g. likes, clicks, shares, saves) to learn user preferences and item characteristics. Beyond these numerical ratings, textual reviews provide insights into users fine-grained preferences and item features. Analyzing these reviews is crucial for enhancing the performance and interpretability of personalized recommendation results. In recent years, review-based recommender systems have emerged as a significant sub-field in this domain. In this paper, we provide a comprehensive overview of the developments in review-based recommender systems over recent years, highlighting the importance of reviews in recommender systems, as well as the challenges associated with extracting features from reviews and integrating them into ratings. Specifically, we present a categorization of these systems and summarize the state-of-the-art methods, analyzing their unique features, effectiveness, and limitations. Finally, we propose potential directions for future research, including the integration of multimodal data, multi-criteria rating information, and ethical considerations.
Paper Structure (46 sections, 14 equations, 5 figures, 6 tables)

This paper contains 46 sections, 14 equations, 5 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Collaborative Filtering
  4. Content-based Filtering
  5. Hybrid Recommender Systems
  6. Why adding review into recommender systems
  7. Review-based recommender systems
  8. Generic Review-based Methods
  9. Aspect-based Methods
  10. Rating and Review Fusion Methods
  11. Ratings and Aspects Fusion Methods
  12. State-of-the-art Review-based Recommender Systems
  13. Probabilistic and Topic Modeling Approaches
  14. Deep Learning-based Approaches
  15. Convolutional Neural Network (CNN)-based Approaches:
  16. ...and 31 more sections

Figures (5)

  • Figure 1: Trends in Review-based Recommender Systems (2015-–2024): Key publications in journals and conferences
  • Figure 2: An example of a hotel review highlighting aspects such as "service", "location" and "cost"
  • Figure 3: A Simple Multi-Layer Perceptron (MLP) with 3 Layers
  • Figure 4: Summary of the Different Review-based Recommender Models
  • Figure 5: The Evolution of the Review-based Recommendation Between 2015 to 2024