Article
Computer Science, Artificial Intelligence
Haoyang Li, Xin Wang, Ziwei Zhang, Jianxin Ma, Peng Cui, Wenwu Zhu
Summary: This paper presents the ISRec method, which aims to capture user intentions to improve recommendation system performance. By extracting intentions from user's historical interaction behaviors and using a message-passing mechanism on an intention graph, the method predicts future user behaviors more accurately and provides transparent and explainable recommendations.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Mingkai He, Jing Lin, Jinwei Luo, Weike Pan, Zhong Ming
Summary: Heterogeneous sequential recommendation is an emerging research topic that deals with item sequences associated with multiple types of feedback. We propose a model called FLAG, which consists of four modules to capture user preferences and make recommendations effectively.
ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Jiajin Wu, Bo Yang, Runze Mao, Qing Li
Summary: Sequential recommendation systems have gained significant attention, but current models still suffer from popularity bias. To alleviate this bias, this study proposes a debiasing model that considers the dynamic user desire and conducts intervention analysis and counterfactual reasoning. The proposed model, PAUDRec, outperforms existing models while alleviating popularity bias in sequential recommendation.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Computer Science, Theory & Methods
Shanming Wei, Shunmei Meng, Qianmu Li, Xiaokang Zhou, Lianyong Qi, Xiaolong Xu
Summary: The ubiquity of recommendation systems has brought significant changes to daily life, but privacy concerns have highlighted the need for innovative solutions. In this study, we propose an edge-enabled federated learning model called KG-FedTrans4Rec that combines knowledge graph information with sequential recommendation tasks. Our experimental results demonstrate the superior performance of our model compared to other recommendation models.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Danyang Wang, Xi Xiong, Yuanyuan Li, Jianghe Wang, Qiurong Tan
Summary: Matching candidate news with user interests is critical for news recommendation. To confront the challenge of multiple user interests, a hierarchical candidate-aware user modeling framework is proposed to accurately match users' multi-field and multi-grained interests with candidate news. Experimental outcomes on large-scale datasets demonstrate the effectiveness and superiority of the proposed method over existing state-of-the-art methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Interdisciplinary Applications
Fan Yang, Gangmin Li, Yong Yue
Summary: Existing recommender systems ignore the users' intentions and driving force, while our proposed model improves recommendation performance by learning user's intentions and preferences, achieving significant improvements.
Article
Computer Science, Information Systems
Honglian Wang, Peiyan Li, Yang Liu, Junming Shao
Summary: This paper proposes a new method for next point-of-interest (POI) recommendation, called DSPR, by exploring user preferences and real-time demand simultaneously to support the final POI recommendation. Experimental results show that DSPR outperforms many state-of-the-art methods in recommendation performance.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Xing Wu, Yisong Li, Jianjia Wang, Quan Qian, Yike Guo
Summary: The recommendation system is widely used in the digital economy to provide personalized services. Capturing the user-item relations efficiently is crucial, but it faces challenges in extracting complicated associations and integrating numerous item connections. To address these challenges, a User Behavior-Aware Recommendation method with knowledge graph (UBAR) is proposed. Experimental results on multiple datasets demonstrate the effectiveness and efficiency of the UBAR method.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Yichi Zhang, Guisheng Yin, Hongbin Dong, Liguo Zhang
Summary: The key problem of sequential recommendation is how to capture user sequential patterns and enrich user sequential representations from historical interactions. Existing methods focus on studying sequential patterns in the time domain, while considering only time domain information may result in difficulties in mining user's sequential patterns. To solve this problem, we propose a novel attention-based frequency-aware multi-scale network that considers both time and frequency domains.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Zhi Liu, Deju Zhang, Chenwei Zhang, Jixin Bian, Junhui Deng, Guojiang Shen, Xiangjie Kong
Summary: Accurate user modeling is crucial for improving user satisfaction with recommended POIs and enriching user experience in point-of-interest (POI) recommendation. However, existing methods lack the ability to capture user-preference knowledge hidden in complex social networks. To address this issue, we propose a novel knowledge-driven and user-aware POI recommendation method called KDRank, which constructs a knowledge graph and employs cross-embedding, knowledge aggregation, and attention mechanism to enhance the accuracy of POI recommendations. Experimental results on real datasets demonstrate the effectiveness of our proposed method.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Chemistry, Multidisciplinary
Zhuoming Xu, Hanlin Liu, Jian Li, Qianqian Zhang, Yan Tang
Summary: This paper proposes a knowledge graph-based recommendation method that captures users' potential interests by utilizing the connections between entities and relations in the knowledge graph. Experimental results show that this method outperforms others in terms of recommendation accuracy and diversity.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Artificial Intelligence
Jihu Wang, Yuliang Shi, Han Yu, Kun Zhang, Xinjun Wang, Zhongmin Yan, Hui Li
Summary: This paper introduces a new sequential recommendation method TDSRec, which improves recommendation performance by incorporating temporal density information and contrastive learning. Extensive experiments on multiple test datasets demonstrate its superior performance.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Ke Sun, Tieyun Qian, Xu Chen, Ming Zhong
Summary: In this paper, we propose a context-aware seq2seq translation model to capture the inter-sequence dependency for sequential recommendations. The injected VAE in our model redresses the semantic imbalance between context and item, leading to superior performance over state-of-the-art baselines.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Zhuoyi Lin, Sheng Zang, Rundong Wang, Zhu Sun, J. Senthilnath, Chi Xu, Chee Keong Kwoh
Summary: In this paper, we propose a novel Intention-aware Re-ranking Model with Dynamic Transformer Encoder (RAISE) to perform user-specific prediction based on their intentions. We mine latent user intentions from text reviews and model them explicitly for each user-item pair. We introduce a dynamic transformer encoder to capture user-specific inter-item relationships, utilizing the learned latent user intentions. Experimental results show the superiority of our proposed RAISE, with significant improvements in Precision@5, MAP@5, and NDCG@5.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Yihu Zhang, Bo Yang, Haodong Liu, Dongsheng Li
Summary: In this paper, a Time-Aware Transformer for Sequential Recommendation (TAT4SRec), an SA-based neural network model that utilizes temporal information to capture users' preferences more precisely, is proposed. TAT4SRec utilizes an encoder-decoder structure to model timestamps and interacted items separately, and different embedding modules are designed to transform continuous data (timestamps) and discrete data (item IDs) into embedding matrices. Experimental results demonstrate the effectiveness of TAT4SRec over various state-of-the-art models and its potential application in online applications.
APPLIED SOFT COMPUTING
(2023)