Article
Engineering, Environmental
Shaodong Zheng, Jinsong Zhao
Summary: With the rapid development of the modern chemical process industry, process monitoring techniques have been investigated to enhance loss prevention capability. In this study, a three-step high-fidelity PU approach based on deep learning is proposed for semi-supervised fault detection of chemical processes. Experimental results demonstrate the effectiveness and superiority of the proposed approach compared to other competing PU learning approaches and supervised fault detection models.
PROCESS SAFETY AND ENVIRONMENTAL PROTECTION
(2022)
Article
Computer Science, Artificial Intelligence
Xueying Shi, Yueming Jin, Qi Dou, Pheng-Ann Heng
Summary: In this paper, a novel two-stage Semi-Supervised Learning method SurgSSL is proposed for label-efficient Surgical workflow recognition, leveraging the inherent knowledge in unlabeled data for improved performance. The method first implicitly excavates motion knowledge from unlabeled data and then performs explicit excavation using pre-knowledge pseudo labeling, achieving superior results compared to existing methods.
MEDICAL IMAGE ANALYSIS
(2021)
Article
Computer Science, Information Systems
Roohallah Alizadehsani, Danial Sharifrazi, Navid Hoseini Izadi, Javad Hassannataj Joloudari, Afshin Shoeibi, Juan M. Gorriz, Sadiq Hussain, Juan E. Arco, Zahra Alizadeh Sani, Fahime Khozeimeh, Abbas Khosravi, Saeid Nahavandi, Sheikh Mohammed Shariful Islam, U. Rajendra Acharya
Summary: This article introduces a semi-supervised classification method using limited labeled data, relying on Sobel edge detection and Generative Adversarial Networks (GANs) to automate COVID-19 diagnosis. Experimental results demonstrate that the proposed method significantly outperforms supervised learning methods in cases where labeled data is scarce.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2021)
Article
Biochemical Research Methods
Raphael Mourad
Summary: Genome-wide association studies have identified thousands of SNPs associated with genetic diseases, but most are in non-coding regions. Predicting molecular processes based on DNA sequence using deep learning has proven effective. To overcome limitations of supervised learning, a shift towards semi-supervised learning utilizing unlabeled sequences is proposed, showing significant performance improvements.
BMC BIOINFORMATICS
(2023)
Article
Computer Science, Artificial Intelligence
Weichao Yi, Liquan Dong, Ming Liu, Mei Hui, Lingqin Kong, Yuejin Zhao
Summary: In this study, a novel Semi-supervised Progressive Dehazing Network (Semi-PDNet) is proposed, which leverages both synthetic and real-world images in the training process. The network follows a progressive architecture with three core stages: image encode stage (IES), feature enhance stage (FES), and hierarchical reconstruction stage (HRS). The stage-by-stage paradigm allows for better haze removal by utilizing informative features from shallow to deep. Additionally, an unlabeled contrastive guidance (UCG) is utilized to bridge the domain gap between synthetic and real-world images.
Article
Computer Science, Artificial Intelligence
Tao Zhang, Tianqing Zhu, Jing Li, Mengde Han, Wanlei Zhou, Philip Yu
Summary: This paper explores the use of semi-supervised learning to address fairness issues in machine learning, including predicting labels for unlabeled data, resampling to obtain multiple fair datasets, and using ensemble learning to improve accuracy and reduce discrimination. Theoretical analysis and experiments demonstrate that this method achieves a better trade-off between accuracy and fairness.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Computer Science, Information Systems
Xiuyi Jia, Tao Wen, Weiping Ding, Huaxiong Li, Weiwei Li
Summary: Label distribution learning (LDL) is a new paradigm in machine learning that addresses label ambiguity by emphasizing the relevance of each label to a particular instance. We propose a projection graph embedding algorithm for semi-supervised label distribution learning (PGE-SLDL), which aims to select valuable features, construct an accurate graph, and recover unknown label distributions. The self-updating projection graph is more effective in learning label distribution compared to traditional fixed graphs in semi-supervised learning.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Alexander Ligthart, Cagatay Catal, Bedir Tekinerdogan
Summary: Opinion spam detection is important for identifying fake reviews. Semi-supervised learning approaches are necessary when data is largely unlabeled. Self-training with Naive Bayes base classifier shows the highest accuracy among tested methods.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Information Systems
Moxian Song, Hongyan Li, Chenxi Sun, Derun Cai, Shenda Hong
Summary: Semi-supervised partial label learning is a learning method that deals with partially labeled and unlabeled data. This paper proposes a novel approach by label set assignment and dependence-maximized dimensionality reduction to obtain reliable label confidences. Extensive experiments validate the effectiveness and superiority of the proposed method.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Cephas A. S. Barreto, Arthur Costa Gorgonio, Joao C. Xavier-Junior, Anne Magaly De Paula Canuto
Summary: Semi-supervised learning (SSL) is a machine learning approach that integrates supervised and unsupervised learning mechanisms. This paper focuses on the use of a wrapper-based strategy in SSL and proposes three selection methods for efficient selection of unlabelled instances. The feasibility of these methods is evaluated through empirical analysis on two well-known SSL methods: Self-training and Co-training.
Article
Computer Science, Artificial Intelligence
Linshan Wu, Leyuan Fang, Xingxin He, Min He, Jiayi Ma, Zhun Zhong
Summary: This paper proposes a semi-supervised semantic segmentation method that learns a segmentation model using limited labeled images and abundant unlabeled images. The approach focuses on generating reliable pseudo labels based on the confidence scores of unlabeled images, but overlooks the use of labeled images. The proposed CISC-R method leverages labeled images to refine the pseudo labels and enhance the performance of semi-supervised semantic segmentation.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Mathematics, Applied
Rongling Lang, Ya Fan, Guoliang Liu, Guodong Liu
Summary: This study proposes a new method for lung sound classification, which can classify respiratory sounds into different categories with a small number of labeled samples and a large number of unlabeled samples, using graph semi-supervised CNN technology, outperforming traditional CNN methods.
APPLIED MATHEMATICS AND COMPUTATION
(2021)
Article
Computer Science, Information Systems
Bo Liu, Junrui Liu, Yanshan Xiao, Qihang Chen, Kai Wang, Ruiguang Huang, Liangjiao Li
Summary: This paper proposes a self-paced algorithm for PU learning that utilizes privileged information and similarity weights to build a more accurate classifier. Experimental results demonstrate its superior performance compared to previous methods.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Ruiqi Guo, Yong Peng, Wanzeng Kong, Fan Li
Summary: In this paper, a semi-supervised Label Distribution Learning model with label Correlations and data Manifold exploration (sLDLCM) is proposed. It effectively utilizes both labeled and unlabeled data to capture the underlying data properties and jointly estimates the label distributions of unlabeled samples and other model variables. Experimental results on multiple tasks demonstrate that the proposed sLDLCM model outperforms the state-of-the-arts.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Zhuo Huang, Jian Yang, Chen Gong
Summary: Semi-Supervised Learning with mismatched classes refers to the problem where the classes-of-interest in labeled data are only a subset of the classes in unlabeled data. To address this, recent methods divide unlabeled data into useful in-distribution (ID) data and harmful out-of-distribution (OOD) data. However, they overlook the potential value of OOD data. Thus, this paper proposes a Transferable OOD data Recycling (TOOR) method to properly utilize both ID data and recyclable OOD data for improved SSL performance.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Computer Science, Artificial Intelligence
Xiaofeng Ding, Yinting Zheng, Zuan Wang, Kim-Kwang Raymond Choo, Hai Jin
Summary: This paper presents a learned spatial textual index for efficiently processing spatial textual data. The index is constructed based on radix table, spline points, and inverted lists, with high-dimensional coordinates converted using Morton encoding. Real-time data insertion, deletion, and update are handled using a gap array and a space reallocation strategy. Query processing algorithms and an optimizer using random forest regression model are proposed to enhance query efficiency. Evaluation results show that the proposed index outperforms the IR-tree in terms of construction time, index size, and query efficiency.
JOURNAL OF INTELLIGENT INFORMATION SYSTEMS
(2023)
Article
Computer Science, Information Systems
Zhiqiu Zhang, Zhu Tianqing, Wei Ren, Ping Xiong, Kim-Kwang Raymond Choo
Summary: In federated learning, a global learning model is trained by the server using gradient information shared by multiple clients to protect client data privacy. However, it has been shown that training data can be reconstructed from shared gradients, leading to privacy breaches. This paper proposes two pruning-based defense mechanisms to prevent privacy leaks in the image reconstruction process and demonstrates their effectiveness on various model architectures and datasets.
COMPUTERS & SECURITY
(2023)
Review
Health Care Sciences & Services
Tongnian Wang, Yan Du, Yanmin Gong, Kim-Kwang Raymond Choo, Yuanxiong Guo
Summary: The proliferation of mHealth applications is driven by technological advancements and the integration of AI. Federated learning (FL) addresses the privacy concerns of sharing raw data by allowing collaborative training without data access. This review highlights FL's potential in mHealth and identifies technical limitations and solutions.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2023)
Article
Computer Science, Hardware & Architecture
Adarsh Kumar, Neelu Jyothi Ahuja, Monika Thapliyal, Sarthika Dutt, Tanesh Kumar, Diego Augusto De Jesus Pacheco, Charalambos Konstantinou, Kim-Kwang Raymond Choo
Summary: Two-thirds of the earth's surface is covered by unexplored water, making underwater monitoring crucial for various purposes such as resource extraction, marine life monitoring, military applications, surveillance, and predicting tidal behavior. Unmanned underwater vehicles and robots are commonly used for exploration due to the complexity of the underwater environment. Blockchain has emerged as an important enabling technology to address security, data sharing, and resource management in underwater applications. This study reviews the utilization of blockchain in underwater applications, discussing use cases, architectures, challenges, solutions, and future research directions.
JOURNAL OF NETWORK AND COMPUTER APPLICATIONS
(2023)
Article
Telecommunications
Biwen Chen, Zhongming Wang, Tao Xiang, Jiyun Yang, Debiao He, Kim-Kwang Raymond Choo
Summary: Vehicular Ad-Hoc Networks (VANETs) have improved driving safety and comfort through vehicular wireless communication technology. However, existing authentication protocols in VANETs have limitations in terms of privacy protection, malicious entity tracking, and cross-domain authentication. To address these challenges, we propose a secure and effective group signature scheme for anonymous authentication and traceable identity within a domain, and a blockchain-based privacy-preserving cross-domain authentication protocol (BCGS) that integrates both blockchain and group signature. Our evaluations show that BCGS outperforms other approaches in terms of security, computation, and storage costs.
VEHICULAR COMMUNICATIONS
(2023)
Article
Engineering, Multidisciplinary
Tao Ye, Min Luo, Yi Yang, Kim-Kwang Raymond Choo, Debiao He
Summary: This survey article focuses on designing redactable blockchain-based solutions. While immutability is essential for blockchain, it can also be misused to disseminate illicit content and violate privacy regulations like GDPR. The article surveys the existing literature on redactable blockchain, discusses limitations and challenges, and highlights future research opportunities.
IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING
(2023)
Article
Computer Science, Information Systems
Milad Taleby Ahvanooey, Mark Xuefang Zhu, Shiyan Ou, Hassan Dana Mazraeh, Wojciech Mazurczyk, Kim-Kwang Raymond Choo, Chuan Li
Summary: Social media platforms have changed how we communicate and share information, but users also willingly share their private sensitive data (PSDs), leading to privacy concerns. This study proposes a novel assessment model to reduce privacy invasion risks of users' PSDs in social media platforms, based on determinant criteria.
COMPUTERS & SECURITY
(2023)
Article
Computer Science, Artificial Intelligence
Henry Chacon, Vishwa Koppisetti, David Hardage, Kim-Kwang Raymond Choo, Paul Rad
Summary: For call center facilities, accurately forecasting the number of call arrivals is crucial for customer satisfaction and budget management. This study compares classical time series methods with machine/deep learning techniques to forecast call arrival, using real-life call logs from a national US insurance company. The results show that deep learning models perform well in short-term periods with enough seasonal data, but boosting approach outperforms all models, including deep learning, in long-term periods. These findings highlight the importance of considering limited seasonality and the use of benchmark approaches for call arrival forecasting.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Xinru Yan, Yinbin Miao, Xinghua Li, Kim-Kwang Raymond Choo, Xiangdong Meng, Robert H. H. Deng
Summary: To address the data island issue in the distributed IoT while preserving privacy, a privacy-preserving federated learning (PPFL) scheme using blockchain is proposed. The scheme tackles the problems of single point of failure and untrusted aggregation results by leveraging blockchain and implements reliable model aggregation in an asynchronous setting using a practical byzantine fault-tolerant protocol. The scheme also improves system robustness by incorporating differential privacy. Security analysis and experiments show that the proposed scheme is secure, robust, and achieves high accuracy compared to state-of-the-art schemes.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Computer Science, Theory & Methods
Lin Chen, Danyang Yue, Xiaofeng Ding, Zuan Wang, Kim-Kwang Raymond Choo, Hai Jin
Summary: In this paper, we propose a method that combines layer-wise relevance propagation with gradient descent to address limitations in deep learning related to data and user privacy. The method injects proper noise into gradients to improve model utility, and uses the NoisyMin algorithm to select the best step size for each gradient perturbation. Experimental evaluations validate the effectiveness of the proposed algorithm and its ability to protect privacy.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2023)
Article
Engineering, Civil
Yinbin Miao, Yutao Yang, Xinghua Li, Kim-Kwang Raymond Choo, Xiangdong Meng, Robert H. Deng
Summary: With the rapid development of Intelligent Transportation System (ITS), a large number of spatial data are generated. However, outsourcing spatial data to the cloud server poses security and privacy issues. This survey aims to summarize advanced privacy-preserving spatial data query schemes and analyze the most advanced solutions. It also compares existing solutions comprehensively and discusses open challenges and potential research directions for privacy-preserving spatial data query.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Theory & Methods
Yinbin Miao, Yutao Yang, Xinghua Li, Zhiquan Liu, Hongwei Li, Kim-Kwang Raymond Choo, Robert H. H. Deng
Summary: With the rapid development of Location-Based Services (LBS), the security issues such as location privacy leakage have become a concern. In this study, an efficient Privacy-preserving Spatial Range Query (PSRQ) scheme is proposed by combining Geohash algorithm with Circular Shift and Coalesce Bloom Filter (CSC-BF) framework and Symmetric-key Hidden Vector Encryption (SHVE). Additionally, a Confused Bloom Filter (CBF) is designed to confuse the inclusion relationship in Bloom filter, and a more secure and practical enhanced scheme PSRQ+ is proposed. The experimental results show significant improvement in query efficiency compared with previous solutions.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2023)
Review
Business
Zainatun N. Hussin, Christie Pei-Yee Chin, Stephen L. Sondoh Jr, Kim-Kwang Raymond Choo
Summary: This article reviews existing studies on government social media use and develops an integrative model to explain the factors influencing its use and impact. The study finds that platform quality, content quality, and government service quality are key predictors of citizen use, mediated by perceived individual benefits and moderated by perceived behavioral control. The article contributes to the field by offering a comprehensive overview and a conceptual model.
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT
(2023)
Article
Business
Christian Baliker, Mohamed Baza, Abdullah Alourani, Ali Alshehri, Hani Alshahrani, Kim-Kwang Raymond Choo
Summary: Financial Technology (FinTech) has expanded its scope from simple mobile banking to include online money transfers, crowdfunding, and managing individual investments. This emphasizes the importance of security and privacy in FinTech, particularly in the use of Blockchain-based FinTech applications. This research systematically summarizes recent developments and highlights the challenges and future directions in Blockchain-based FinTech applications.
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT
(2023)
Article
Engineering, Civil
Keke Gai, Haokun Tang, Guangshun Li, Tianxiu Xie, Shuo Wang, Liehuang Zhu, Kim-Kwang Raymond Choo
Summary: This paper highlights the importance of data-driven applications in modern maritime transportation systems, particularly in communication and safety decision-making. The authors propose using blockchain to protect privacy and ensure data accuracy, and evaluate the security and performance of their approach.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)