Article
Computer Science, Information Systems
Batyr Charyyev, Mehmet Hadi Gunes
Summary: Engineered systems are becoming smarter thanks to computing capabilities and the increasing number of IoT devices. However, IoT devices are vulnerable to compromise due to their limited resources, making them prime targets for malicious activities. This article introduces a novel approach using locality-sensitive hash to identify IoT devices based on their traffic flow, achieving high precision and recall without the need for feature extraction or model retraining. The evaluation results demonstrate that this approach performs on par with state-of-the-art machine learning-based methods.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Engineering, Multidisciplinary
Lianyong Qi, Xiaokang Wang, Xiaolong Xu, Wanchun Dou, Shancang Li
Summary: Recommender systems use collaborative filtering to return a list of recommended items that best match the user preferences, but face challenges in integrating QoS data securely and balancing data availability with user privacy. By enhancing LSH technique, a method for cross-platform recommendation decision-making is proposed and experiments show its advantages over other competing methods.
IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Huawen Liu, Wenhua Zhou, Hong Zhang, Gang Li, Shichao Zhang, Xuelong Li
Summary: This research introduces a novel hash bit reduction schema to derive shorter binary codes, effectively reducing the number of hash bits and improving the retrieval performance of LSH.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Software Engineering
Xiaoxiao Chi, Chao Yan, Hao Wang, Wajid Rafique, Lianyong Qi
Summary: With the increasing variety and volume of web services in the IoT age, this article proposes a unique amplified locality-sensitive hashing (LSH)-based service recommendation method, SRAmplified-LSH, to address the challenges of user privacy and efficient recommendation.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2022)
Article
Computer Science, Software Engineering
Hongsheng Hu, Gillian Dobbie, Zoran Salcic, Meng Liu, Jianbing Zhang, Lingjuan Lyu, Xuyun Zhang
Summary: Recommender systems are important in big data analytics for their potential to bring high profit. However, privacy concerns and regulations make it difficult to integrate scattered data. Existing privacy-preserving recommender system models based on cryptography lack flexibility. This paper proposes a differentially private LSH approach that guarantees privacy preservation while offering efficient and accurate recommendations.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2023)
Article
Engineering, Mechanical
Emma Lejeune, Peerasait Prachaseree
Summary: Living systems have an amazing ability to sense, store, and respond to mechanical stimuli, and there is increasing interest in designing engineered systems to replicate this functionality. This work focuses on the question of whether mechanical systems can transform mechanical information into sensor readouts to meet the requirements for a locality sensitive hash function. The findings suggest that different mechanical systems vary in their effectiveness in performing this task, and this research serves as a starting point for future investigation into designing and optimizing mechanical systems for conveying mechanical information for downstream computing.
EXTREME MECHANICS LETTERS
(2023)
Article
Biochemical Research Methods
Wontack Han, Haixu Tang, Yuzhen Ye
Summary: Microbial organisms are crucial for human health and diseases. Computational and machine learning methods have been developed to analyze microbial features and predict host phenotypes. This study improves the subtractive assembly approach and identifies differential genes using k-mers with similar abundance profiles.
JOURNAL OF COMPUTATIONAL BIOLOGY
(2022)
Article
Computer Science, Cybernetics
Wenqiang He, Yongjun Li, Yinyin Zhang, Xiangyu Li
Summary: In this paper, a novel approach is proposed to address the problem of cross-site user identification. By incorporating locality-sensitive hashing technology and approximate nearest neighbors searching strategy, the proposed method reduces computation and improves efficiency.
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Amira Jouirou, Abir Baazaoui, Walid Barhoumi
Summary: This paper introduces an efficient MVIF-CBMR method based on late fusion, which combines retrieval result-level of MLO and CC views. By adopting a coupled multi-index with a dynamic distance, the method improves the retrieval performance by fully exerting the discriminative power of the complementary MLO-CC features.
PATTERN RECOGNITION
(2021)
Article
Automation & Control Systems
Wajahat Ali, Ikram Ud Din, Ahmad Almogren, Mohsen Guizani, Mansour Zuair
Summary: The smart grid integrates advanced technologies to monitor and present energy usage in real-time, improving customer satisfaction. Protecting the privacy of industrial ecosystems requires more attention, while also considering reducing computational resources, communication overhead, and ensuring data accuracy.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Computer Science, Information Systems
Ameerah Abdullah Alshahrani, Emad Sami Jaha
Summary: With the advancement of multimedia technology, the use of large image libraries has expanded. Image retrieval, especially for face images, has become crucial in many applications such as social media. This study proposes a locality-sensitive hashing (LSH) based method to improve the performance of face image retrieval from large-scale databases. The experimental results show that the proposed method outperforms the traditional method, achieving higher accuracy in less time.
Article
Computer Science, Information Systems
Chao Liu, Zengxi Li, Shunshun Liu, Jushi Xie, Chao Yan, Wanli Huang
Summary: With the growing popularity of sports items and businesses, player transfer events have become more frequent, creating a big market with hidden transfer fees. To address concerns like privacy and fairness, a trusted player transfer evaluation method called TPTELSH+B is proposed, using Locality-Sensitive Hashing (LSH) and Blockchain technologies. Experimental results show that TPTELSH+B has good evaluation performance compared to other approaches.
Article
Computer Science, Artificial Intelligence
Hanrui Zhang, Qianmu Li, Jiangmin Xu, Shunmei Meng, Jun Hou
Summary: This study proposes a privacy-preserving collaborative filtering recommendation method with clustering and locality-sensitive hashing to address the problems of data sparseness, cold start, and privacy protection in intelligent recommender systems. The method predicts and fills missing ratings in sub-rating matrices, combines them into a complete rating matrix, and uses locality-sensitive hashing to quickly obtain neighbors for the target user. Experimental results show improved accuracy and privacy protection.
COMPUTATIONAL INTELLIGENCE
(2022)
Article
Computer Science, Information Systems
Chintoo Kumar, C. Ravindranath Chowdary, Deepika Shukla
Summary: This paper investigates the automatic detection of group formation using locality-sensitive hashing to achieve ordered and flexible preferences in group recommendation. By applying the MinHash technique and locality-sensitive hashing, similar users can be efficiently identified and automatically formed into groups, making the recommendation system more satisfying for users. Experimental results show that the proposed model improves consensus among users in a group.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Hardware & Architecture
Hanwen Liu, Shunmei Meng, Jun Hou, Shuo Wang, Qianmu Li, Chanying Huang
Summary: This paper proposes a potential social relationships prediction approach based on locality-sensitive hashing (LSH), which clusters similar users and predicts the types of social relationships among them using a fuzzy computing method. The rationality of the prediction results is verified using the social balance theory.
JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS
(2022)
Article
Computer Science, Software Engineering
Hongsheng Hu, Gillian Dobbie, Zoran Salcic, Meng Liu, Jianbing Zhang, Lingjuan Lyu, Xuyun Zhang
Summary: Recommender systems are important in big data analytics for their potential to bring high profit. However, privacy concerns and regulations make it difficult to integrate scattered data. Existing privacy-preserving recommender system models based on cryptography lack flexibility. This paper proposes a differentially private LSH approach that guarantees privacy preservation while offering efficient and accurate recommendations.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2023)
Article
Automation & Control Systems
Yang Xu, Md Zakirul Alam Bhuiyan, Tian Wang, Xiaokang Zhou, Amit Kumar Singh
Summary: In this article, we propose a framework called C-fDRL to protect the context-aware privacy of task offloading using context-aware federated deep reinforcement learning. The framework operates in three stages (CloudAI, EdgeAI, and DeviceAI) of the overall system, decoupling data from tasks through a context-aware data management approach for local and edge computation, leading to improved data privacy protection.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Computer Science, Information Systems
Liang Wang, Zhiwen Yu, Kaishun Wu, Dingqi Yang, En Wang, Tian Wang, Yihan Mei, Bin Guo
Summary: Mobile Crowdsensing (MCS) is an appealing paradigm for collaboratively collecting data from surrounding environments by assigning outsourced sensing tasks to volunteer workers. However, unpredictable disruptions during task implementation often result in task execution failure and impair the benefit of MCS systems. In this work, we propose a robust task assignment scheme that proactively creates assignments offline, aiming to strengthen the robustness of the scheme and minimize workers' traveling detour cost. By leveraging workers' spatiotemporal mobility, we construct an assignment graph and use an evolutionary multi-tasking optimization algorithm (EMTRA) to achieve adequate Pareto-optimal schemes. Comprehensive experiments on real-world datasets validate the effectiveness and applicability of our approach.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Computer Science, Information Systems
Jing Bai, Zhiwen Zeng, Tian Wang, Shaobo Zhang, Neal N. Xiong, Anfeng Liu
Summary: In this article, a trust-based active notice task offloading (TANTO) scheme is proposed to provide trust and low-delay task offloading for resource-limited IoT devices in areas with no available communication infrastructure. The main innovations of TANTO include a novel task offloading mechanism, a trust calculation and reasoning method, and an online UAV trajectory optimization algorithm. Experimental results show that TANTO outperforms previous studies in terms of task completion rate, tasks' average completion time, and UAV's flight cost.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Computer Science, Information Systems
Xuyun Zhang, Lianyong Qi, Wanchun Dou, Qiang He, Christopher Leckie, Ramamohanarao Kotagiri, Zoran Salcic
Summary: Scalable data processing platforms built on cloud computing are becoming more attractive for supporting big data applications. However, privacy concerns hinder the use of public cloud platforms. In this paper, a scalable approach based on MapReduce is proposed to address the scalability issues of multidimensional anonymisation when handling big data. The approach also demonstrates applicability to differential privacy.
IEEE TRANSACTIONS ON BIG DATA
(2022)
Article
Computer Science, Theory & Methods
Guoming Zhang, Xuyun Zhang, Muhammad Bilal, Wanchun Dou, Xiaolong Xu, Joel J. P. C. Rodrigues
Summary: With the rapid growth of medical costs, controlling medical expenses has become an important task for the Health Insurance Department. Traditional per-service payment methods in medical insurance lead to many unreasonable expenses. To address this issue, the use of single-disease payment mechanisms has become popular, but it also carries the risk of fraud. This study proposes a framework based on consortium blockchain and deep learning to identify fraud in medical insurance, automating the recognition of suspicious medical records and ensuring valid implementation of single-disease payments, while reducing the workload of medical insurance auditors.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2022)
Article
Computer Science, Information Systems
Jiwei Zhang, Md Zakirul Alam Bhuiyan, Xu Yang, Tian Wang, Xuesong Xu, Thaier Hayajneh, Faiza Khan
Summary: Internet of Things (IoT) is a rapidly developing technology that is vulnerable to cyberattacks and anomalies. This article proposes a framework called AntiConcealer, which uses edge artificial intelligence (EdgeAI) to detect adversary concealed behaviors in the IoT. The usability and reliability of the framework are verified through evaluation using honeypots integrated with edge servers.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Computer Science, Information Systems
Xiyuan Hu, Jingpeng Sun, Jinping Dong, Xuyun Zhang
Summary: We proposed a deep learning-based accurate snore detection model for long-term home monitoring of snoring during sleep. The model outperformed other traditional approaches and deep learning models in terms of snore detection. It predicted candidate boxes and confidence scores based on the feature maps derived by the feature extraction network.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2023)
Article
Engineering, Electrical & Electronic
Lianyong Qi, Xiaolong Xu, Xiaotong Wu, Qiang Ni, Yuan Yuan, Xuyun Zhang
Summary: This paper investigates the opportunities and challenges of mobile video streaming services in the sixth generation (6G) network using digital twin technology and cloud-centric architecture. The authors propose a solution using crowdsourcing to attract mobile users to follow a specified path and share network resources, alleviating the problem of increasing traffic demands. They present algorithms for user recruitment optimization and evaluate the system performance through extensive experiments.
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS
(2023)
Editorial Material
Environmental Sciences
Carlos Enrique Montenegro Marin, Xuyun Zhang, Nallappan Gunasekaran
JOURNAL OF THE INDIAN SOCIETY OF REMOTE SENSING
(2023)
Article
Computer Science, Information Systems
Soubhagya Ranjan Mallick, Rakesh K. Lenka, Veena Goswami, Suraj Sharma, Asish Kumar Dalai, Himansu Das, Rabindra K. Barik
Summary: Recent research in healthcare systems has focused on integrating IoT, Blockchain, and cloud computing to improve IoT device performance, create smart healthcare platforms, and provide optimal healthcare services. Data collection, processing, geolocation, access management, device prioritization, and storage are primary challenges in IoT healthcare systems. The use of Blockchain technology in healthcare platforms offers data privacy, anonymity, and validity. This paper presents a novel decentralized Blockchain-enabled geospatial service architecture called BCGeo for smart healthcare systems, providing online geospatial healthcare services for newcomers in Bhubaneswar, India. The framework includes immutability, scalability, geospatial mapping, patient prioritization, and decentralized privacy protection policies, addressing technical challenges in current healthcare systems.
Article
Computer Science, Hardware & Architecture
Yi Chen, Zoran Salcic, Hongxia Wang, Kim-Kwang Raymond Choo, Xuyun Zhang
Summary: This article proposes a joint distortion-based non-additive cost assignment method to reduce distortion drift and improve security in video steganography. Extensive experiments show that the proposed method achieves enhanced security and visual stego video quality.
IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Lianyong Qi, Wenmin Lin, Xuyun Zhang, Wanchun Dou, Xiaolong Xu, Jinjun Chen
Summary: Using Web APIs registered in service sharing communities for mobile APP development can reduce development period and cost. However, the large number and differences of available APIs make it difficult for selection. To address this challenge, a correlation graph-based approach is proposed for personalized and compatible Web APIs recommendation in mobile APP development. Extensive experiments on a real dataset prove the feasibility of the proposed recommendation approach.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Proceedings Paper
Computer Science, Information Systems
Shuhui Fan, Shaojing Fu, Yuchuan Luo, Haoran Xu, Xuyun Zhang, Ming Xu
Summary: The rapid growth of cryptocurrency market value has attracted more investors, but the anonymity of blockchain also makes it a tool for criminals. Smart contract scam detection is crucial for investors, and traditional methods perform poorly in detecting scams with similar codes.
PROCEEDINGS OF THE 31ST ACM INTERNATIONAL CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT, CIKM 2022
(2022)
Proceedings Paper
Computer Science, Information Systems
Wei Shen, Xiaonan He, Chuheng Zhang, Xuyun Zhang, Jian Xie
Summary: Spoken dialogue systems face challenges in responding accurately due to noisy input, semantic misunderstandings, or lack of knowledge. To improve the user experience, researchers propose a proactive interaction mechanism that predicts user satisfaction and asks clarification questions before providing a response. They develop a pipeline to predict user satisfaction using weak labels and a transformer-based model, and introduce a contextual user satisfaction metric to evaluate the experience. Their model deployed on DuerOS shows a relative improvement in user satisfaction prediction accuracy and user experience.
PROCEEDINGS OF THE 31ST ACM INTERNATIONAL CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT, CIKM 2022
(2022)