Article
Computer Science, Information Systems
Chuan Lin, Wei Zhou, Junhao Wen
Summary: This paper proposes an interest-aware contrastive-learning-based GCN model (IC-GCN), which can effectively utilize signals from higher-order neighbors in recommender systems and address some issues in GCN models. Experimental results demonstrate the effectiveness of the IC-GCN model.
Article
Computer Science, Artificial Intelligence
Guofang Ma, Yuexuan Wang, Xiaolin Zheng, Xiaoye Miao, Qianqiao Liang
Summary: This paper proposes a novel Trust-aware Latent Space Mapping approach (TLSM-CDR) for Cross-domain Recommendation, addressing the challenge of insufficient bridged users by considering users' trust relationships. Experimental results demonstrate that the TLSM-CDR model significantly outperforms several state-of-the-art methods on two real-world datasets.
Article
Engineering, Multidisciplinary
Liangtian Wan, Feng Xia, Xiangjie Kong, Ching-Hsien Hsu, Runhe Huang, Jianhua Ma
Summary: Recent years have seen a surge in information overload on online social networks, leading to increased interest in social network based recommender systems. This study introduces a novel trust-aware approach based on deep learning to improve recommendation performance, incorporating deep matrix factorization techniques, deep marginalized denoising autoencoder, and community regularization, which outperformed existing baselines, particularly for cold-start users.
IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Le Wu, Xiangnan He, Xiang Wang, Kun Zhang, Meng Wang
Summary: Influenced by the success of deep learning, research in recommendation has shifted to developing new recommender models based on neural networks. This survey paper systematically reviews neural recommender models from the perspective of recommendation modeling with the accuracy goal, aiming to summarize the field and facilitate researchers and practitioners. It categorizes the work into collaborative filtering, content enriched recommendation, and temporal/sequential recommendation based on the data usage, and discusses promising directions in the field.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Milad Ahmadian, Mahmood Ahmadi, Sajad Ahmadian
Summary: This paper proposes a trust-aware recommendation method based on deep sparse autoencoder to address challenges in deep learning based recommendation systems. Through the use of an effective probabilistic model and an implicit rating utilization mechanism, the method achieves significant improvements in generating latent features and providing accurate recommendations.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Jung-Hsien Chiang, Chung-Yao Ma, Chi-Shiang Wang, Pei-Yi Hao
Summary: Recommendation systems have become more important due to the abundance of information on the internet. This study proposes a novel recommendation system that adapts to users' changing preferences by using context factors and attention mechanisms. Experimental results show that this context-aware recommendation model outperforms traditional methods.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Information Systems
Jie Nie, Zian Zhao, Lei Huang, Weizhi Nie, Zhiqiang Wei
Summary: Recently, cross-domain recommendation has been proposed to handle the data sparsity and cold-start problems in recommendation systems, which have been widely used in online business scenarios. However, most existing methods ignore user-group information and rely on matrix factorization for generating embeddings, resulting in weak generalization ability of user latent features. In this paper, a novel cross-domain recommendation model called UCMF is proposed, which enhances user representation learning using user clustering and multidimensional information fusion. Experimental results demonstrate that UCMF outperforms state-of-the-art methods in cross-domain recommendation on real-world datasets.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Review
Computer Science, Artificial Intelligence
Wu-Dong Xi, Ling Huang, Chang-Dong Wang, Yin-Yu Zheng, Jian-Huang Lai
Summary: Many recommender systems utilize review text as auxiliary information to enhance recommendation quality, but existing models typically use ratings as the ground truth for error backpropagation, potentially resulting in the loss of valuable review information. This article introduces a novel deep model DRRNN, which considers both target ratings and reviews as ground truth for error backpropagation, allowing for the retention of more semantic information in rating predictions. Extensive experiments on four publicly available datasets demonstrate the effectiveness of the proposed DRRNN model in terms of rating prediction.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Information Systems
Chao Wu, Sannyuya Liu, Zeyu Zeng, Mao Chen, Adi Alhudhaif, Xiangyang Tang, Fayadh Alenezi, Norah Alnaim, Xicheng Peng
Summary: This paper proposes a knowledge graph-based multi-context-aware recommendation algorithm for learning user/item representations. It automatically discovers representative user preference templates and learns high-order connectivity between long-distance user-item pairs.
INFORMATION SCIENCES
(2022)
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
Asma Sattar, Davide Bacciu
Summary: This paper proposes a context-aware Graph Convolutional Matrix Completion method, which integrates structural information, user opinions, and surrounding context to generate personalized recommendations. The effectiveness of the model is demonstrated through experiments on 14 datasets.
NEURAL PROCESSING LETTERS
(2023)
Article
Computer Science, Artificial Intelligence
Tingting Liang, Lin Ma, Weizhong Zhang, Haoran Xu, Congying Xia, Yuyu Yin
Summary: In this paper, a novel neural network called DHGCN is proposed for item recommendation. By considering the interactions between users and items, users and users, and items and items, and using heterogeneous graph convolution for representation aggregation, efficient and accurate recommendation is achieved.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Bo Chen, Yue Ding, Xin Xin, Yunzhe Li, Yule Wang, Dong Wang
Summary: AIRec is an attentive intersection model for TRS that constructs user representations through multi-layer perceptron and hierarchical attention network, enhances learning by exploiting intersection between user and item tags, and significantly improves tag-aware top-n recommendation.
Article
Computer Science, Information Systems
Chunting Wei, Jiwei Qin, Wei Zeng
Summary: Most existing list ranking methods focus on learning either linear or nonlinear interactions, while ignoring the coexistence and complementary roles of both. To address this issue, the proposed DNR framework combines traditional and deep learning methods to jointly learn linear and nonlinear features for improved recommendation system performance.
Article
Computer Science, Artificial Intelligence
Mengqi Zhang, Shu Wu, Meng Gao, Xin Jiang, Ke Xu, Liang Wang
Summary: This paper proposes a method called A-PGNN, which is a personalized graph neural network with attention mechanism, to address the problem of session-aware recommendation. A-PGNN outperforms existing methods by extracting personalized structural information and modeling the effects of historical sessions explicitly.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Business, Finance
Yingli Wang, Chang Lu, Xiaoguang Yang, Qingpeng Zhang
Summary: This study empirically analyzes the impact of China's Manufacturing Purchasing Managers' Index (PMI) announcements on the stock market, and finds that positive PMI announcements have a significant effect while negative announcements do not. Furthermore, the effect of positive announcements becomes increasingly significant during economic expansion and stability periods. This finding contradicts international experience and may be explained by the behavior of individual investors in China's stock market.
INTERNATIONAL JOURNAL OF FINANCE & ECONOMICS
(2023)
Article
Cardiac & Cardiovascular Systems
Sharen Lee, Jiandong Zhou, Keith Sai Kit Leung, Abraham Ka Chung Wai, Kamalan Jeevaratnam, Emma King, Tong Liu, Wing Tak Wong, Carlin Chang, Ian Chi Kei Wong, Bernard Man Yung Cheung, Gary Tse, Qingpeng Zhang
Summary: Based on real-world data of type 2 diabetic patients in Hong Kong, the use of SGLT2Is was associated with lower risk of incident AF, stroke/transient ischemic attack, and cardiovascular and all-cause mortality outcomes compared to DPP4Is.
CARDIOVASCULAR DRUGS AND THERAPY
(2023)
Article
Computer Science, Information Systems
Xin Wang, Wenqi Hao, Yuzhou Qin, Baozhu Liu, Pengkai Liu, Yanyan Song, Qingpeng Zhang, Xiaofei Wang
Summary: The paper proposes a novel indexing solution, FPIRPQ, that accelerates RPQ query processing by leveraging frequent path mining. The researchers also design a micro-benchmark to evaluate the performance of this approach.
WORLD WIDE WEB-INTERNET AND WEB INFORMATION SYSTEMS
(2023)
Article
Management
Lifang Li, Hong Wen, Qingpeng Zhang
Summary: This study examines the characteristics of situational information sharing on Weibo and WeChat during the Changsheng fake vaccine crisis in China. The findings reveal that WeChat is more focused on sharing notifications, cautionary advice, and criticism, while Weibo is more inclined towards emotional support and help-seeking information. The study provides valuable insights for authorities and researchers in improving crisis communication and public emergency management.
JOURNAL OF CONTINGENCIES AND CRISIS MANAGEMENT
(2023)
Article
Mathematics, Applied
Weibin Cheng, Hanchu Zhou, Yang Ye, Yifan Chen, Fengshi Jing, Zhidong Cao, Daniel Dajun Zeng, Qingpeng Zhang
Summary: The accumulation of susceptible populations for respiratory infectious diseases when COVID-19-targeted non-pharmaceutical interventions were in place might pose a greater risk of future outbreaks. The timing and magnitude of respiratory infectious disease resurgence after lifting COVID-19-targeted non-pharmaceutical interventions were examined, along with the burdens on the health system. The proposed Threshold-based Control Method (TCM) identifies effective dynamic non-pharmaceutical interventions and ensures a sufficient supply of hospital beds for severely infected patients.
Article
Computer Science, Information Systems
Wenlong Liu, Jiahua Pan, Xingyu Zhang, Xinxin Gong, Yang Ye, Xujin Zhao, Xin Wang, Kent Wu, Hua Xiang, Houmin Yan, Qingpeng Zhang
Summary: Product matching aims to identify identical or similar products sold on different platforms. This paper introduces a two-stage pipeline to match products from eBay and Amazon. A new framework called RAEA is employed for fine filtering, which focuses on the interactions between attribute and relation triples.
WORLD WIDE WEB-INTERNET AND WEB INFORMATION SYSTEMS
(2023)
Article
Computer Science, Information Systems
Lifang Li, Jiandong Zhou, Jun Zhuang, Qingpeng Zhang
Summary: Understanding the effects of gender-specific emotional responses on information sharing behaviors is crucial for effective public health crisis communication. This study examines anxiety and anger levels among males and females during the Changsheng vaccine crisis and the COVID-19 pandemic, and their impact on the virality of crisis information. The findings reveal significant gender differences in emotional responses, with high anxiety and anger levels among females. The study also highlights the influence of gender-specific emotional expressions on crisis information, especially for male influencers.
INFORMATION PROCESSING & MANAGEMENT
(2023)
Article
Computer Science, Interdisciplinary Applications
Chaocheng He, Fuzhen Liu, Ke Dong, Jiang Wu, Qingpeng Zhang
Summary: Studying research leadership relations and constructing a research leadership network can help us understand the formation mechanisms of research collaborations. The results show that research leadership relations are reciprocal and based on a local hierarchy. Cognitive and institutional proximity have a significant influence on tie formation.
JOURNAL OF INFORMETRICS
(2023)
Article
Engineering, Civil
Yan Zhang, Keyang Sun, Di Wen, Dingjun Chen, Hongxia Lv, Qingpeng Zhang
Summary: In this study, a method based on spatiotemporal convolutional neural network (STCNN) is proposed for short-term passenger flow forecasting of origin-destination (OD) pairs in urban rail transit. The method constructs the spatial and temporal relationships among critical OD pairs and learns their features using convolutional and temporal convolutional neural networks. Experimental results on a field dataset show that the proposed STCNN outperforms state-of-the-art methods in accurately predicting passenger flows for critical OD pairs.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Information Systems
Yao Yao, Hanchu Zhou, Zhidong Cao, Daniel Dajun Zeng, Qingpeng Zhang
Summary: This study predicts the epidemic curves of respiratory infectious diseases (RID) after COVID-19 and develops optimal adaptive nonpharmaceutical interventions (NPIs) strategies using a deep reinforcement learning (DRL) model to minimize health and economic costs. The results show that the DRL-based adaptive NPIs successfully suppress the outbreak of both COVID-19 and RID, achieving herd immunity in a short period without overwhelming hospitals. The insights from this study can be extended to other countries/regions.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2023)
Article
Computer Science, Information Systems
Ping Liang, Jiannan Yang, Weilan Wang, Guanjie Yuan, Min Han, Qingpeng Zhang, Zhen Li
Summary: Early diagnosis and prediction of CKD progress are crucial for personalized treatment and improving patients' quality of life. This study explores the intelligibility of machine learning and deep learning models for ESRD prediction in CKD patients. The deep learning model achieves high accuracy and provides intelligible insights into CKD progression. This study provides solid data-driven evidence for using machine learning in CKD clinical management and treatment.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2023)
Editorial Material
Psychology, Biological
Qingpeng Zhang
Summary: Online health communities play a crucial role in providing valuable social support in China, which cannot be matched by artificial intelligence.
NATURE HUMAN BEHAVIOUR
(2023)
Article
Immunology
Xue Liang, Jiming Li, Yuan Fang, Qingpeng Zhang, Martin C. S. Wong, Fuk-yuen Yu, Danhua Ye, Paul Shing-fong Chan, Joseph Kawuki, Siyu Chen, Phoenix K. H. Mo, Zixin Wang
Summary: This study investigated the associations of COVID-19 vaccination, perceptions related to COVID-19 and seasonal influenza vaccination (SIV), with the behavioural intention to receive SIV among older adults in Hong Kong, China. It found that concerns about the negative impact of SIV and COVID-19 vaccination on each other and the perceived risk of co-infection with COVID-19 and seasonal influenza were associated with the intention to receive SIV.
Article
Computer Science, Cybernetics
Tianyi Luo, Duo Xu, Zhidong Cao, Pengfei Zhao, Jiaojiao Wang, Qingpeng Zhang
Summary: This article proposes a multilayer network-based model for the coupling transmission of information, behavior, and disease. It considers various factors such as psychological drivers of information dissemination, the impact of herd mentality on behavioral transmission, and the dynamics of the COVID-19 Omicron mutant strain and relevant countermeasures. The model was leveraged to assess the effects of propagation parameters and interlayer coupling parameters on the magnitude of the COVID-19 outbreak and strain on medical resources, providing valuable perspectives for epidemic prevention strategies.
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
(2023)
Article
Health Policy & Services
Fengshi Jing, Qingpeng Zhang, Weiming Tang, Johnson Zixin Wang, Joseph Tak-fai Lau, Xiaoming Li
Summary: This study develops a data-driven approach to reconstruct the social network of MSM communities. The method accurately reconstructs the social networks using locally observed information and parameter learning.
AIDS CARE-PSYCHOLOGICAL AND SOCIO-MEDICAL ASPECTS OF AIDS/HIV
(2023)