Multi-objective optimization with recommender systems: A systematic review

Zaizi, Fatima Ezzahra; Qassimi, Sara; Rakrak, Said

doi:10.1016/j.is.2023.102233

Cited by 12 publications

(2 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By incorporating multi-objective optimization into contextual bandit algorithms, it becomes possible to consider and optimize for diverse objectives such as relevance, diversity, fairness, novelty, and serendipity in the recommendation process. Scalarization and population-based heuristics are widely used techniques in multiobjective optimization and decision-making [29]. Scalarization is employed to convert a multi-objective problem (MOP) into a single-objective problem (SOP), allowing the utilization of existing optimization methods developed for SOPs using methods such as weighting methods [30], lexicographic [31], e-constraint [32], or Tchebycheff methods [33], etc.…”

Section: Related Workmentioning

confidence: 99%

Multi-Objective Contextual Bandits in Recommendation Systems for Smart Tourism

Qassimi,

Rakrak

2024

Preprint

Self Cite

View full text Add to dashboard Cite

In the context of smart tourism, the utilization of recommender systems is becoming increasingly critical for enhancing the personalization and quality of travel experiences. Tourists often encounter complex decision-making due to information overload, context-aware recommender systems have emerged as a promising solution, leveraging contextual data such as time, weather, and location. However, these systems face the challenge of the complexity of handling dynamic context. Thus, the static nature of these systems can result in a degradation of performance, as they fail to capture the dynamic nature of user behavior and context. Addressing these issues, this paper presents a novel multi-objective contextual multi-armed bandit-based recommender system. This proposal integrates the strengths of contextual bandit algorithms with multi-objective optimization, offering personalized recommendations and learning from user feedback. The multi-objective optimization includes the dual necessities of relevance and fairness in recommendations, ensuring the promotion of a balanced tourism ecosystem. Extensive experiments were carried out on public datasets to evaluate the performance of our proposed approach. Its effectiveness was compared with baseline methods to establish its performance, demonstrating the significance of multi-objective optimization in enhancing personalized recommendations in smart tourism. To evaluate the performance of our proposed algorithm, we conducted experiments using two datasets, a designed dataset that simulates real-world scenarios and TripAdvisor dataset. The study provides a case scenario of implementing this proposed approach in the smart tourism context of Marrakesh, demonstrating its potential to revolutionize the tourist experience in smart cities.

show abstract

Section: Related Workmentioning

confidence: 99%

Multi-Objective Contextual Bandits in Recommendation Systems for Smart Tourism

Qassimi,

Rakrak

2024

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Multi-task learning (MTL) has emerged as a promising direction, improving generalization by leveraging useful information across related tasks [5,6]. By jointly optimizing rating prediction, item preference classification, and auxiliary tasks, MTL models can effectively fuse heterogeneous feedback signals, leading to more accurate and robust recommendations.…”

Section: Introductionmentioning

confidence: 99%

Enhancing Personalized Recommendations: A Study on the Efficacy of Multi-Task Learning and Feature Integration

Wang,

Jin,

Zhang

et al. 2024

Information

View full text Add to dashboard Cite

Personalized recommender systems play a crucial role in assisting users in discovering items of interest from vast amounts of information across various domains. However, developing accurate personalized recommender systems remains challenging due to the need to balance model architectures, input feature combinations, and fusion of heterogeneous data sources. This study investigates the impacts of these factors on recommendation performance using the MovieLens and Book Recommendation datasets. Six models, including single-task neural networks, multi-task learning, and baselines, were evaluated with various input feature combinations using Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). The multi-task learning approach achieved significantly lower RMSE and MAE by effectively leveraging heterogeneous data sources for personalized recommendations through a shared neural network architecture. Furthermore, incorporating user data and content data progressively enhanced performance compared to using only item identifiers. The findings highlight the importance of advanced model architectures and fusing heterogeneous data sources for high-quality recommendations, providing valuable insights for designing effective recommender systems across diverse domains.

show abstract