Article
Computer Science, Artificial Intelligence
Runlong Yu, Qi Liu, Yuyang Ye, Mingyue Cheng, Enhong Chen, Jianhui Ma
Summary: In this paper, a new framework called Collaborative List-and-Pairwise Filtering (CLAPF) is proposed, which aims to introduce pairwise thinking into listwise methods. By smoothing rank-biased measure and combining multiple metrics to capture the performance of top-k recommendation, as well as discussing a sampling scheme to accelerate convergence speed, our empirical studies demonstrate that CLAPF outperforms state-of-the-art approaches on real-world datasets.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Computer Science, Information Systems
Anchen Li, Bo Yang, Huan Huo, Farookh Khadeer Hussain
Summary: A novel recommendation method is proposed in this study, which enhances recommendation performance by mining implicit relations between users and items. Experimental results demonstrate that the method achieves superior performance in rating prediction and Top-k recommendation.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Interdisciplinary Applications
Frans Prathama, Wenny Franciska Senjaya, Bernardo Nugroho Yahya, Jei-Zheng Wu
Summary: Recommendation systems are crucial for assisting users in finding relevant items by constructing user profiles and learning their preferences. The use of implicit feedback can improve the recommendation quality but poses challenges in interpretability and diverse interpretability due to different characteristics.
COMPUTERS & INDUSTRIAL ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Jia-Lue Chen, Jia-Jia Cai, Yuan Jiang, Sheng-Jun Huang
Summary: The article discusses a method to improve model performance in recommender systems by treating the task as a Positive Unlabeled learning problem. The active learning approach proposed can significantly reduce labeling cost while achieving superior performance in multiple criteria.
NEURAL PROCESSING LETTERS
(2021)
Article
Computer Science, Information Systems
Runzhi Xu, Jianjun Li, Guohui Li, Peng Pan, Quan Zhou, Chaoyang Wang
Summary: The recommender system is crucial in dealing with data explosion, and the application of deep neural networks has become a popular research topic. Methods like SDNN and DualCF have improved the efficiency of capturing user-item relations in recommendation systems, and their effectiveness has been verified through extensive experiments.
INFORMATION SCIENCES
(2022)
Article
Chemistry, Analytical
Fei Pan, Xiaoyu Zhao, Boda Zhang, Pengjun Xiang, Mengdie Hu, Xuesong Gao
Summary: In this paper, a neural network-based CSI feedback model, Mix_Multi_TransNet, is proposed. It achieves higher accuracy in both indoor and outdoor scenes by considering the spatial characteristics and temporal sequence of the channel, as well as reducing the number of model parameters.
Article
Computer Science, Artificial Intelligence
Xiaocong Chen, Lina Yao, Julian McAuley, Guanglin Zhou, Xianzhi Wang
Summary: This article provides a timely and comprehensive overview of recent trends in deep reinforcement learning (DRL) in recommender systems. It discusses the motivation for applying DRL in recommender systems, presents a taxonomy and summary of current DRL-based recommender systems, and explores emerging topics and open issues. The survey serves as an introductory material for readers from academia and industry and identifies notable opportunities for further research.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Chemistry, Multidisciplinary
Thi-Linh Ho, Anh-Cuong Le, Dinh-Hong Vu
Summary: Recommender systems face the challenge of providing accurate recommendations that cater to the diverse preferences of users. Most studies primarily use the utility matrix and do not integrate textual sources with it. To overcome this challenge, we propose a novel method that efficiently integrates textual and utility matrix information using the Transformer Model. Experimental results demonstrate that our model outperforms baseline models (MLP), as well as SVD and graph-based models, in terms of recommendation accuracy.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Xuchen Xia, Wenming Ma, Jinkai Zhang, En Zhang
Summary: Graph collaborative filtering is an efficient method for finding hidden user interests in recommender systems. However, data sparsity poses a challenge for recommender systems. To address this, researchers have proposed a novel method called CECL, which combines contrastive learning and community detection to utilize latent information in the data, resulting in improved performance compared to existing methods.
Article
Computer Science, Artificial Intelligence
Pratik K. Biswas, Songlin Liu
Summary: In this paper, a hybrid recommender system that combines collaborative filtering with deep learning is proposed to enhance recommendation performance and overcome the limitations of collaborative filtering. By combining the outputs of collaborative filtering with a deep neural network in a big data processing framework, the proposed system outperforms existing hybrid recommender systems in recommending smartphones to prospective customers.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Lei Chen, Guixiang Zhu, Weichao Liang, Youquan Wang
Summary: Trip recommendation is an intelligent service that offers personalized itinerary plans to tourists in unfamiliar cities, considering temporal and spatial constraints. In this article, we propose MORL-Trip, a Multi-Objective Reinforcement Learning approach, to address the challenges of capturing users' dynamic preferences and enhancing the diversity and popularity of personalized trips. MORL-Trip models the recommendation as a Markov Decision Process and incorporates sequential, geographic, and order information to learn user's context. It also introduces a composite reward function to reinforce accuracy, popularity, and diversity as principal objectives.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Telecommunications
Guanghui Fan, Jinlong Sun, Guan Gui, Haris Gacanin, Bamidele Adebisi, Tomoaki Ohtsuki
Summary: Due to the lack of channel reciprocity in FDD massive MIMO systems, downlink CSI needs to be continuously fed back to the base station from the user equipment, consuming bandwidth resources. This paper proposes a fully convolutional neural network for compressing and decompressing the downlink CSI. Experimental results demonstrate that the proposed method outperforms the baseline in terms of reconstruction performance and reduction of storage and computational overhead, and is robust to quantization error in real feedback scenarios.
IEEE TRANSACTIONS ON COGNITIVE COMMUNICATIONS AND NETWORKING
(2022)
Article
Automation & Control Systems
Zhiwei Guo, Heng Wang
Summary: This article proposes a social recommendation framework based on deep graph neural networks for future IoT recommendation systems. It encodes user and item feature spaces and completes missing values in user-item rating matrices through matrix factorization. Experiments confirm the efficiency and stability of the proposed framework.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Automation & Control Systems
Chang-Dong Wang, Yan-Hui Chen, Wu-Dong Xi, Ling Huang, Guangqiang Xie
Summary: The study introduces a novel neural network model CEICFNet to address the sparsity and cold-start issues in recommender systems. By learning latent factors across domains, the model effectively integrates explicit ratings and implicit interactions.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2022)
Article
Computer Science, Information Systems
Seungyeon Lee, Dohyun Kim
Summary: In this paper, a recommender system based on convolutional neural network is proposed to capture the complex interactions between users and items, giving greater weight to important features and alleviate the overfitting issue. Experiments show that the proposed method outperforms existing methods.
INFORMATION SCIENCES
(2022)