Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Comprehensive Review of Recommender Systems: Transitioning from Theory to Practice

Shaina Raza, Mizanur Rahman, Safiullah Kamawal, Armin Toroghi, Ananya Raval, Farshad Navah, Amirmohammad Kazemeini

TL;DR

This survey assesses the progression of Recommender Systems from foundational theory to practical deployment between 2017 and 2024. It organizes methods into foundational, deep-learning, graph-based, sequential, knowledge-based, RL, LLM-based, and multimodal families, and covers specialized domains such as context-aware, review-based, aspect-based, and fairness-driven RS. The paper highlights challenges across domains, including scalability, privacy, fairness, explainability, and data sparsity, and discusses how modern approaches like GNNs, LLMs, SSL, and multimodal fusion address them. It also maps RS applications across e-commerce, entertainment, health, tourism, finance, and beyond, and advocates stronger academia-industry collaboration, responsible AI practices, and evaluation beyond accuracy to guide future research and deployment efforts.

Abstract

Recommender Systems (RS) play an integral role in enhancing user experiences by providing personalized item suggestions. This survey reviews the progress in RS inclusively from 2017 to 2024, effectively connecting theoretical advances with practical applications. We explore the development from traditional RS techniques like content-based and collaborative filtering to advanced methods involving deep learning, graph-based models, reinforcement learning, and large language models. We also discuss specialized systems such as context-aware, review-based, and fairness-aware RS. The primary goal of this survey is to bridge theory with practice. It addresses challenges across various sectors, including e-commerce, healthcare, and finance, emphasizing the need for scalable, real-time, and trustworthy solutions. Through this survey, we promote stronger partnerships between academic research and industry practices. The insights offered by this survey aim to guide industry professionals in optimizing RS deployment and to inspire future research directions, especially in addressing emerging technological and societal trends\footnote. The survey resources are available in the public GitHub repository https://github.com/VectorInstitute/Recommender-Systems-Survey. (Recommender systems, large language models, chatgpt, responsible AI)

A Comprehensive Review of Recommender Systems: Transitioning from Theory to Practice

TL;DR

This survey assesses the progression of Recommender Systems from foundational theory to practical deployment between 2017 and 2024. It organizes methods into foundational, deep-learning, graph-based, sequential, knowledge-based, RL, LLM-based, and multimodal families, and covers specialized domains such as context-aware, review-based, aspect-based, and fairness-driven RS. The paper highlights challenges across domains, including scalability, privacy, fairness, explainability, and data sparsity, and discusses how modern approaches like GNNs, LLMs, SSL, and multimodal fusion address them. It also maps RS applications across e-commerce, entertainment, health, tourism, finance, and beyond, and advocates stronger academia-industry collaboration, responsible AI practices, and evaluation beyond accuracy to guide future research and deployment efforts.

Abstract

Recommender Systems (RS) play an integral role in enhancing user experiences by providing personalized item suggestions. This survey reviews the progress in RS inclusively from 2017 to 2024, effectively connecting theoretical advances with practical applications. We explore the development from traditional RS techniques like content-based and collaborative filtering to advanced methods involving deep learning, graph-based models, reinforcement learning, and large language models. We also discuss specialized systems such as context-aware, review-based, and fairness-aware RS. The primary goal of this survey is to bridge theory with practice. It addresses challenges across various sectors, including e-commerce, healthcare, and finance, emphasizing the need for scalable, real-time, and trustworthy solutions. Through this survey, we promote stronger partnerships between academic research and industry practices. The insights offered by this survey aim to guide industry professionals in optimizing RS deployment and to inspire future research directions, especially in addressing emerging technological and societal trends\footnote. The survey resources are available in the public GitHub repository https://github.com/VectorInstitute/Recommender-Systems-Survey. (Recommender systems, large language models, chatgpt, responsible AI)
Paper Structure (64 sections, 10 equations, 3 figures, 6 tables)

This paper contains 64 sections, 10 equations, 3 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Historical Context
  4. Current State of Practice and Theory in Recommender Systems
  5. Literature Review Methodology
  6. Research Questions
  7. Databases Searched
  8. Search Query
  9. Quality Assessment
  10. Challenges in Recommender Systems
  11. E-commerce
  12. Entertainment
  13. News
  14. Tourism
  15. Healthcare
  16. ...and 49 more sections

Figures (3)

  • Figure 1: Overview of Publication Trends and Key Venues in RS Research.
  • Figure 2: General Framework of RS showing an outline of a data processing pipeline. It covers stages from data acquisition, preparation, and feature engineering, through model training and validation, to deployment and monitoring. The pipeline also includes steps for candidate generation, ranking, evaluation, and a feedback loop to ensure continuous improvement and business value.
  • Figure 3: Timeline of Foundational Recommender Systems.