Article
Computer Science, Hardware & Architecture
Ru Huang, Lei Ma, Jianhua He, Xiaoli Chu
Summary: The passage introduces the definition of complex networks and their applications in real systems, proposes T-GAN as a new method for predicting temporal complex networks, and describes the framework and application instances of the method.
Article
Computer Science, Artificial Intelligence
Xinrui Dong, Yijia Zhang, Kuo Pang, Fei Chen, Mingyu Lu
Summary: In this paper, a heterogeneous graph neural network with denoising (HGNND) is proposed to address the issue of meta-path selection in graph embedding. By projecting the features of different types of nodes into a common vector space and utilizing a graph neural network to aggregate neighbor node information and capture the structure information of the heterogeneous graph, the proposed model achieves state-of-the-art performance on three real-world datasets.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Zezhi Shao, Yongjun Xu, Wei Wei, Fei Wang, Zhao Zhang, Feida Zhu
Summary: Graph neural networks have been widely used for heterogeneous graph embedding due to their ability to effectively encode rich information. However, previous methods often fail to fully utilize the heterogeneity and semantics in complex local structures. To address this issue, the authors propose MV-HetGNN, which comprehensively learns complex heterogeneity and semantics to generate versatile node representations.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Lijun Dong, Hong Yao, Dan Li, Yi Wang, Shengwen Li, Qingzhong Liang
Summary: Graph embedding technique in artificial intelligence is important for processing complex graph data efficiently. Existing GNN models often have limitations in considering global topology information, leading to difficulties in distinguishing nodes with similar local topologies. The proposed AS-GNN model aims to address this issue by capturing global topology information based on the characteristics of complex networks.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Dongyue Chen, Ruonan Liu, Qinghua Hu, Steven X. Ding
Summary: This article proposes a novel interaction-aware and data fusion method, named interaction-aware graph neural networks (IAGNNs), for fault diagnosis of complex industrial processes. It can learn multiple interactions between sensor signals and extract fault features, demonstrating reliable and superior performance in fault diagnosis.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Yangrui Chen, Jiaxuan You, Jun He, Yuan Lin, Yanghua Peng, Chuan Wu, Yibo Zhu
Summary: This paper proposes a new class of GNNs called SP-GNNs, which enhance the expressive power of GNN architectures by incorporating a position encoder and a structure encoder. The experiments show significant improvement in classification using SP-GNNs on various graph datasets.
Article
Computer Science, Artificial Intelligence
Yongjing Hao, Jun Ma, Pengpeng Zhao, Guanfeng Liu, Xuefeng Xian, Lei Zhao, Victor S. Sheng
Summary: Graph neural networks (GNNs) have been widely used in recommendation systems due to their advantages in graph representation learning, and many successful models have applied graph-based methods for sequential recommendation. However, existing research only considers the number of interactions between items, neglecting the multi-dimensional transformation relationships. Therefore, we propose a Category and Time information integrated Graph Neural Network (CT-GNN) that combines item category and interaction time information to form fine-grained item representations, and design a temporal self-attention network for dynamic user preference modeling and next-item recommendation. Experimental results on real-world datasets demonstrate the excellent performance of the proposed model. (c) 2023 Elsevier Ltd. All rights reserved.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
Shengyingjie Liu, Zongkai Yang, Sannyuya Liu, Ruxia Liang, Jianwen Sun, Qing Li, Xiaoxuan Shen
Summary: Intelligent tutoring systems (ITS) analyze user behavior to customize personalized learning strategies. However, existing methods cannot effectively model ITS data due to the discrete user evolution. This study introduces the concept of a discrete evolution graph (DEG) and proposes the DEGE method to embed ITS data in a hyperbolic space, outperforming other baselines in question annotation and performance prediction.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Computer Science, Artificial Intelligence
Tongya Zheng, Zunlei Feng, Tianli Zhang, Yunzhi Hao, Mingli Song, Xingen Wang, Xinyu Wang, Ji Zhao, Chun Chen
Summary: This study proposes a method called transition propagation graph neural networks (TIP-GNN) to tackle the challenges of encoding nodes' transition structures. The TIP-GNN approach encodes transition structures through multistep transition propagation and distills information from neighborhoods through bilevel graph convolution, resulting in improved accuracy in temporal link prediction.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Di Jin, Zhizhi Yu, Dongxiao He, Carl Yang, Philip S. Yu, Jiawei Han
Summary: Heterogeneous information network (HIN) embedding is a research area that aims to map the structure and semantic information in a HIN to distributed representations. Current graph neural network methods often use a hierarchical attention structure to capture information from meta-path-based neighbors, but this structure is not effective in selecting meta-paths and does not distinguish direct relationships from indirect ones. To address these issues, we propose a novel neural network method that implicitly utilizes attention and meta-paths to alleviate overfitting in HIN.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Dengdi Sun, Dashuang Li, Zhuanlian Ding, Xingyi Zhang, Jin Tang
Summary: This paper proposes a novel all-to-all graph autoencoder model, named A2AE, for multi-view graph representation learning. It utilizes the rich relational information in multiple views and recognizes the importance of different views.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Dengdi Sun, Dashuang Li, Zhuanlian Ding, Xingyi Zhang, Jin Tang
Summary: The study introduces a dual-decoder graph autoencoder model that effectively embeds the topological structure and node attributes of a graph into a compact representation, showcasing superior performance in experiments.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Jing Liu, Tongya Zheng, Qinfen Hao
Summary: This paper proposes a high-order relational knowledge distillation framework for extracting relational knowledge from pre-trained heterogeneous graph neural networks. Experimental results demonstrate that the framework significantly improves prediction performance and possesses good effectiveness and generalization ability.
Article
Engineering, Industrial
Pai Zheng, Liqiao Xia, Chengxi Li, Xinyu Li, Bufan Liu
Summary: This study introduces a multi-agent reinforcement learning method based on industrial knowledge graph to achieve a Self-X cognitive manufacturing network. By establishing IKG, performing embedding algorithm, and implementing a decentralized system, it enables self-configurable solution searching, task decomposition, and self-optimization.
JOURNAL OF MANUFACTURING SYSTEMS
(2021)
Review
Chemistry, Analytical
Van Thuy Hoang, Hyeon-Ju Jeon, Eun-Soon You, Yoewon Yoon, Sungyeop Jung, O-Joun Lee
Summary: Graphs are effective data structures for representing relational data. Graph representation learning is important for various downstream tasks. It aims to map graph entities to low-dimensional vectors while preserving graph structure and relationships. This paper provides a comprehensive overview of graph representation learning models, including traditional and state-of-the-art models on different graphs. It discusses various types of embedding models, practical applications, and challenges for existing models and future research directions.
Article
Computer Science, Information Systems
Usman Ahmed, Jerry Chun-Wei Lin, Gautam Srivastava, Alireza Jolfaei
Summary: In this article, the authors studied adversarial evasion attacks in an active learning environment and proposed a feature subset selection method to prevent evasion attacks in IoT environments. They trained an independent classification model for individual Android applications by extracting application-specific data. By comparing and evaluating Android malware benchmarks using ensemble-based active learning, followed by the use of a collaborative machine learning classifier, they demonstrated protection against adversarial evasion attacks. The proposed approach achieved a receiver operating characteristic of 0.91 with 14 fabricated input features.
IEEE SYSTEMS JOURNAL
(2023)
Article
Computer Science, Artificial Intelligence
Jia-Hao Syu, Jerry Chun-Wei Lin, Chi-Jen Wu, Jan-Ming Ho
Summary: This article introduces a stock selection system called TripleS, which is based on fuzzy set theory. It utilizes the position size to extract the suitability index (SI) that describes the characteristics of stocks and their suitability for certain strategies. Experimental results show that TripleS has minor improvement in precision but significant improvement in investment performance. TripleS-EFT, one of the proposed methods, achieves outstanding profitability and outperforms the benchmark system and state-of-the-art fuzzy-based system in terms of annual return and Sharpe ratio.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Engineering, Civil
Yibing Liu, Lijun Huo, Jun Wu, Ali Kashif Bashir
Summary: As city boundaries expand and vehicles increase, the transportation system faces increasing overload, which has negative effects on people's commutes and overall work and life. However, with the development of 6G-driven Intelligent Transportation Systems (ITS), it becomes possible to alleviate urban congestion. Existing solutions are limited in their ability to optimize traffic efficiently. Therefore, we propose the Direction Decide as a Service (DDaaS) scheme, which incorporates a novel three-layer service architecture, improved modeling and aggregation methods, and a dynamic traffic control algorithm to effectively reduce traffic congestion.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Automation & Control Systems
Phu Pham, Loan T. T. Nguyen, Ngoc-Thanh Nguyen, Witold Pedrycz, Unil Yun, Jerry Chun-Wei Lin, Bay Vo
Summary: This article presents a novel approach called SemHE4Rec, which is a semantic-aware heterogeneous information network (HIN) embedding-based recommendation model. Two embedding techniques are used to learn representations of users and items, and are combined with matrix factorization (MF) for better recommendation performance. Experimental results demonstrate that the proposed SemHE4Rec outperforms recent state-of-the-art HIN embedding-based recommendation techniques.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Chemistry, Multidisciplinary
Atul Kumar Verma, Rahul Saxena, Mahipal Jadeja, Vikrant Bhateja, Jerry Chun-Wei Lin
Summary: Graph Neural Networks (GNNs) have made significant progress in processing graph datasets, and Graph Convolutional Networks (GCNs) have outperformed other models in tasks such as node classification, link prediction, and graph classification. A novel training technique based on network centrality is proposed in this paper, which improves the performance of GCN and GAT models. Empirical analysis shows that the proposed technique achieves better classification accuracy compared to existing methods, and it sets new records on benchmark datasets.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Kakali Chatterjee, Ashish Singh, Neha, Keping Yu
Summary: The quality of healthcare is crucial for users, and smart healthcare systems utilize IoT devices to capture and share patient data for quality services. However, this exposes the system to privacy threats. In this paper, a ring signature-based authentication scheme is proposed to protect the privacy of doctors and patients during collaborative medical consultation, ensuring network quality with the new KMOV Cryptosystem.
ACM JOURNAL OF DATA AND INFORMATION QUALITY
(2023)
Article
Computer Science, Information Systems
Usman Ahmed, Jerry Chun-Wei Lin, Gautam Srivastava
Summary: Cyber-manufacturing Systems (CMS) are an innovative manufacturing model that emphasizes innovation, automation, better customer service, and intelligent systems. They can improve efficiency and productivity, revolutionize product production, and have a potential impact on society and industry. This study comprehensively analyzes CMS and identifies important issues and research opportunities associated with them.
ACM TRANSACTIONS ON INTERNET TECHNOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Wei Fang, Chongyang Li, Qiang Zhang, Xin Zhang, Jerry Chun-Wei Lin
Summary: This paper proposes an efficient biobjective evolutionary algorithm (BOEA-FHUI) for obtaining frequent and high utility itemsets (FHUIs) from transactional databases. The algorithm reduces the search space, improves the search efficiency, and finds better results through pruning, repair, and an improved mutation strategy based on the sparse nature of FHUI.
APPLIED SOFT COMPUTING
(2023)
Review
Chemistry, Analytical
Usman Tariq, Irfan Ahmed, Ali Kashif Bashir, Kamran Shaukat
Summary: The emergence of IoT technology has brought both vast possibilities and new vulnerabilities to connected systems. Developing a secure IoT ecosystem requires a systematic approach to identify and mitigate security threats, with cybersecurity research playing a critical role. The primary challenge is defending against both known and unknown attacks, and concerns regarding connectivity, communication, and management protocols need to be addressed. This research paper provides a comprehensive review of current IoT security concepts, analyzing prevalent security concerns and establishing security goals for specific IoT use cases.
Article
Telecommunications
Yu Zhang, Guoming Tang, Qianyi Huang, Yi Wang, Kui Wu, Keping Yu, Xun Shao
Summary: Non-intrusive load monitoring (NILM) disaggregates a household's main electricity consumption to individual appliance usages, reducing the cost of load monitoring and promoting green homes. Federated learning (FL) can address the privacy concern in NILM applications, but faces challenges of edge resource restriction, model personalization, and training data scarcity. We propose FedNILM, a FL paradigm for edge clients, which utilizes collaborative data aggregation, cloud model compression, and personalized edge model building to provide privacy-preserving and personalized NILM services. Experiments on real-world energy data demonstrate that FedNILM achieves accurate personalized energy disaggregation while protecting user privacy.
IEEE TRANSACTIONS ON GREEN COMMUNICATIONS AND NETWORKING
(2023)
Article
Telecommunications
Zhiwei Guo, Keping Yu, Kostromitin Konstantin, Shahid Mumtaz, Wei Wei, Peng Shi, Joel J. P. C. Rodrigues
Summary: With the development of green wireless communication, the green Internet of Vehicles (GIoV) has emerged as a potential solution for future transportation. Intelligent traffic forecasting for key nodes in GIoV is a significant research topic. This work combines deep embedding and graph embedding to propose a deep collaborative intelligence-driven traffic forecasting model in GIoV, aiming to establish more reliable feature spaces and improve forecasting efficiency. Experimental results on a real-world dataset show a reduction in forecasting deviation by about 15%-25%.
IEEE TRANSACTIONS ON GREEN COMMUNICATIONS AND NETWORKING
(2023)
Article
Computer Science, Information Systems
Peng He, Chunhui Lan, Ali Kashif Bashir, Dapeng Wu, Ruyan Wang, Rupak Kharel, Keping Yu
Summary: In this article, a three-layer federated learning architecture is proposed to reduce training latency and improve efficiency in medical image classification while ensuring data privacy.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2023)
Article
Computer Science, Information Systems
Lei Liu, Yuxing Tian, Chinmay Chakraborty, Jie Feng, Qingqi Pei, Li Zhen, Keping Yu
Summary: This paper introduces a federated learning-based intelligent traffic flow forecasting model, which integrates spatial-temporal graph-based deep learning with the Multilevel Federated Learning framework. The proposed model tackles the issues of centralized data and privacy leakage, while considering both spatial and semantic dependencies to improve prediction accuracy.
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT
(2023)
Article
Computer Science, Information Systems
Fan Yang, Jie Huang, Arpit Bhardwaj, Amir Hussain, Ahmed A. Abd El-Latif, Keping Yu
Summary: The article proposes a nondata-aided error vector (NDA-EVM) technique for evaluating channel quality and symbol error rate (SER) in smart systems. Based on NDA-EVM, an adaptive modulation strategy is designed to address the low spectral efficiency caused by big data transmission.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Computer Science, Cybernetics
Zhiwei Guo, Qin Zhang, Feng Ding, Xiaogang Zhu, Keping Yu
Summary: This article proposes a novel fake news detection model for the context of mixed languages using a multiscale transformer to capture the semantic information of the text. Experimental results show that the proposed method outperforms commonly used baseline models by 2%-10% in terms of detection accuracy.
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
(2023)