Article
Computer Science, Information Systems
Anwar Shah, Bahar Ali, Masood Habib, Jaroslav Frnda, Inam Ullah, Muhammad Shahid Anwar
Summary: This article introduces a three-way decision mechanism called E3FRM, which utilizes human visual characteristics to improve accuracy and trust in visual-based explainable human-computer interaction systems. Experimental results show that E3FRM outperforms existing methods in F1, Accuracy, and Recall by up to 12.8%, 9.6%, and 13.9%, respectively. Therefore, this model has the potential to enhance face recognition accuracy and trust in machines.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2023)
Article
Computer Science, Information Systems
Chengyong Jin, Bao Qing Hu
Summary: This paper introduces the concept of hesitant sets to unify different types of fuzzy sets. It discusses the construction of decision evaluation functions and three-way decisions based on hesitant sets in three-way decision spaces. The paper presents methods for constructing decision evaluation functions in semi-three-way and quasi-three-way decision spaces, as well as transformation methods to three-way decision spaces. The importance of these methods is supported by numerous examples.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Information Systems
A. Savchenko
Summary: A novel image recognition algorithm based on sequential three-way decisions is introduced to speed up the inference in a convolutional neural network. This approach does not require a special training procedure for neural networks and can be used with arbitrary architectures, demonstrating a reduction in running time of up to 40% with a controlled decrease in accuracy when tested on several datasets and neural architectures.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Information Systems
Shahzad Khan, Omar Khan, Nouman Azam, Ihsan Ullah
Summary: Spectral clustering is an unsupervised machine learning algorithm that groups similar data points into clusters by modeling pair-wise data points and utilizing spectral properties. It is well suited for solving problems involving complex patterns. However, it is sensitive to outliers, measurement errors, or perturbations in the original data, which can affect its performance. In this article, a three-way decision based approach is proposed to make spectral clustering insensitive to noise, and it outperforms classical spectral clustering by an average of 30% on various standard datasets.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Information Systems
Yi Xu, Min Wang, Fan Luo
Summary: The movement strategy is a crucial issue that aims to transfer objects from unfavorable regions to favorable regions. Object-based movement strategy is complex but can successfully move many objects, while region-based movement strategy is simple but can only move a few objects. In order to design a movement strategy with lower complexity and higher success rate, a cluster-based movement strategy is proposed, which divides objects into clusters and assigns movement rules to each cluster based on defined criteria.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Information Systems
Yongjing Zhang, Guannan Li, Wangchen Dai, Chengxin Hong, Jin Qian, Zhaoyang Han
Summary: In the context of IoT, decision models and GRC can be used for data processing. By incorporating device-free sensor data into decision models, IoT data can be processed more efficiently and accurate location positioning can be achieved.
COMPUTER COMMUNICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Yusheng Li, Xueyan Shao
Summary: Crime linkage is a challenging task with significant importance in maintaining social security. This paper proposes a method for automatically learning thresholds of three-way decisions without the need for explicit loss functions. Multiple traditional classification algorithms are employed to evaluate the effectiveness of the proposed method, and the results show a reduction in classification errors.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Qinghua Zhang, Xuechao Zhi, Yongyang Dai, Guoyin Wang
Summary: This paper proposes a democratic three-way decision method based on a voting mechanism, which obtains decision opinions from different attributes at a coarse granularity level to improve decision results and optimize their rationality using normalized information gain ratio. Experimental results demonstrate that this method has higher accuracy and better comprehensive evaluation index compared to other methods.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2022)
Article
Chemistry, Analytical
Shiqi Wang, Suen Guan, Hui Lin, Jianming Huang, Fei Long, Junfeng Yao
Summary: Micro-expressions are rapid and subtle facial movements that are difficult to detect and recognize. Micro-expression recognition has attracted extensive attention from computer vision due to its potential applications in various domains. In this article, the authors propose an OF-PCANet+ method for micro-expression recognition, which combines a shallow PCANet+ model with optical flow sequence stacking to learn discriminative spatiotemporal features. Experimental results on publicly available datasets demonstrate that the lightweight model outperforms traditional hand-crafted methods and achieves comparable performance with deep learning based methods.
Article
Biochemistry & Molecular Biology
Jaroslav Malina, Hana Kostrhunova, Peter Scott, Viktor Brabec
Summary: Fe(II)-based metallohelices were found to stabilize various DNA junctions, with the highest selectivity for the Y-shaped 3WJ. These metallohelices were shown to efficiently kill cancer cells and induce DNA damage, offering potential therapeutic benefits.
NUCLEIC ACIDS RESEARCH
(2023)
Article
Computer Science, Information Systems
Yi Xu, Baofeng Li
Summary: This paper discusses the importance of multiview and multilevel in granular computing, and introduces a new partition order product space model. It proposes search algorithms and fusion strategies for solving three-way decisions from multiple views and multiple levels.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Yi Xu, Zhiqiang Zheng, Xiao Liu, Aiting Yao, Xuejun Li
Summary: In mobile edge computing, deploying three-way decisions for service migration can prevent wrong decisions and improve performance. Users are categorized into different regions based on their movement trajectory, and corresponding operations are executed accordingly.
INFORMATION SCIENCES
(2022)
Article
Engineering, Electrical & Electronic
Kun-Hong Liu, Qiu-Shi Jin, Huang-Chao Xu, Yee-Siang Gan, Sze-Teng Liong
Summary: This paper proposes a method to improve the performance of micro-expression recognition systems by introducing genetic algorithms, and experiments show that its performance surpasses other state-of-the-art methods.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2021)
Article
Automation & Control Systems
Xiang Li, Hai Wang, Zeshui Xu
Summary: The study proposes a novel three-way decision method for making decisions on whether enterprises should resume work post-epidemic. The method involves describing enterprise attributes, calculating attribute weights using the entropy weight method, and ultimately making decision results based on minimizing losses.
INTERNATIONAL JOURNAL OF FUZZY SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Longjun Yin, Qinghua Zhang, Fan Zhao, Dun Liu, Guoyin Wang
Summary: This paper analyzes the superiority of three-way decisions (3WDs) over traditional two-way decisions (2WDs) and proposes a 3WDs model based on confidence level and sample mean. Experiments show that this model can effectively reduce the cost of information acquisition and has similar classification accuracy to the 2WDs model.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Engineering, Electrical & Electronic
Linghua Zhou, Weidong Min, Deyu Lin, Qing Han, Ruikang Liu
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2020)
Article
Computer Science, Information Systems
Deyu Lin, Weidong Min, Jianfeng Xu
IEEE INTERNET OF THINGS JOURNAL
(2020)
Article
Computer Science, Information Systems
Deyu Lin, Quan Wang, Weidong Min, Jianfeng Xu, Zhiqiang Zhang
ACM TRANSACTIONS ON SENSOR NETWORKS
(2020)
Article
Computer Science, Artificial Intelligence
Jingwen Duan, Weidong Min, Deyu Lin, Jianfeng Xu, Xin Xiong
Summary: A Multimodal Graph Inference Network (MGIN) is proposed in this study to improve the inference capability of triplets, especially for uncommon samples, by incorporating prior statistical knowledge and fusing visual and semantic features. The method achieves higher average recall and mean recall compared with state-of-the-art methods, with significant improvements in predicting relationships with low probability.
APPLIED INTELLIGENCE
(2021)
Article
Computer Science, Information Systems
Deyu Lin, Lilei Gao, Weidong Min
Summary: In this study, the social welfare theory is used to promote energy balance in the wireless sensor networks by controlling Cluster Head election. A novel Energy-efficient Cluster Head Election scheme is proposed and verified to be effective in improving energy efficiency and prolonging the network lifespan.
IEEE SYSTEMS JOURNAL
(2021)
Article
Computer Science, Hardware & Architecture
Deyu Lin, Weidong Min, Jianfeng Xu, Jiaxun Yang, Jianlin Zhang
Summary: This letter introduces a novel routing method to improve energy efficiency among different clusters, and verifies its effectiveness through extensive experiments.
IEEE EMBEDDED SYSTEMS LETTERS
(2021)
Article
Engineering, Electrical & Electronic
Deyu Lin, Linghe Kong, Chengkun Zhao, Jiayi Gao, Hao Ouyang, Ziyuan Yang, Zhiqiang Zhang
Summary: This paper proposes the novel concept of EIec to evaluate energy consumption equality and establishes related theorems for selecting cluster heads. Based on this, the paper proposes an energy-efficiency-adaptive cluster formation mechanism using economic theory and conducts extensive experiments to assess its effectiveness in improving energy efficiency and network performance.
IET COMMUNICATIONS
(2022)
Article
Computer Science, Information Systems
Qing Han, Huiting Liu, Weidong Min, Tiemei Huang, Deyu Lin, Qi Wang
Summary: This paper proposes a person re-identification method based on 3D skeleton and two-stream approach, which can solve the problems in person re-identification and outperforms existing methods in recognition accuracy according to experimental results.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Chengkun Zhao, Qian Wu, Deyu Lin, Zhiqiang Zhang, Yujie Zhang, Linghe Kong, Yong Liang Guan
Summary: This paper investigates the issue of energy balance in wireless sensor networks and proposes schemes of energy-balanced unequal clustering and energy-efficient cluster head rotation. Through theoretical derivation and mathematical calculation, the optimal number of cluster heads for energy consumption balance is obtained based on the concept of gradient. A fuzzy logic-based mechanism for cluster head rotation is also proposed to balance the energy distribution. Finally, extensive simulations demonstrate the effectiveness of the proposed approach in reducing energy consumption and prolonging network lifetime.
Article
Chemistry, Multidisciplinary
Jingjing Guo, Li Fu, Qian Wang, Yuling Peng, Yaoyao Yuan, Deyu Lin, Yuehui Sun, Hui Wang, Tong Guo
Summary: The interactions between bovine serum albumin (BSA) and four lactic acid-based deep eutectic solvents (LADESs) were investigated using spectroscopic and electrochemical methods. The binding constants, binding sites, and quenching constants were determined, revealing static quenching. Hydrogen bonds and van der Waals forces were found to play a role in the interaction between LADESs and BSA. The energy transfer efficiency and binding distance were calculated, and the quenching of fluorescence was attributed to energy transfer. The study also identified the ideal LADES extractant for BSA separation and purification, and investigated changes in the alpha-helix content of BSA after interaction with LADESs using circular dichroism spectroscopy.
NEW JOURNAL OF CHEMISTRY
(2023)
Article
Computer Science, Artificial Intelligence
Chan Su, Jianguo Wei, Deyu Lin, Linghe Kong
Summary: This article introduces a novel method for facial expression recognition, which focuses on discriminative attention regions and pretrains on ImageNet to alleviate over-fitting. Experimental results demonstrate the superior performance of this method on multiple benchmark datasets.
PATTERN ANALYSIS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Deyu Lin, Zihao Lin, Linghe Kong, Yong Liang Guan
Summary: This paper proposes an energy-efficient routing protocol based on Constrained Minimum Spanning Tree (CMSTR) to balance the energy consumption for intra-cluster communications and solve the problem of overlong intra-cluster communication paths. The simulation experiments show that CMSTR greatly prolongs the network lifetime and has excellent performance in terms of energy efficiency and network stability.
Article
Computer Science, Information Systems
Hao Chen, Xixiang Lv, Wei Zheng, Deyu Lin
Summary: In this article, the VerFHS framework is proposed to address the issue of image privacy protection in the scenario of IoT and cloud computing. By extending the secure k-NN algorithm, cleverly designing ciphertext inner product, and utilizing blockchain reward mechanism, the framework achieves verifiability, feedback, and high security. Furthermore, the VerFHSD scheme is demonstrated to achieve forward security and the effectiveness of the scheme is verified through experiments.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Guanghua Long, Deyu Lin, Jie Lei, Zhiyong Guo, Yangyang Hu, Linglin Xia
Summary: Social bots, which exist widely in major social networks, can be maliciously used to manipulate public opinion, steal user privacy, and spread rumors, posing a serious security threat. This paper proposes a method for detecting malicious social bots by combining sentiment features. It utilizes a Bidirectional Long Short-Term Memory model with an Attention Mechanism to perform sentiment analysis on the online text of social accounts, and analyzes the sentiment fluctuations to derive new features. The new features, along with metadata features, are then input into different machine learning models for analysis and comparison, resulting in improved detection accuracy.
2022 5TH INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND NATURAL LANGUAGE PROCESSING, MLNLP 2022
(2022)
Article
Computer Science, Information Systems
Luchuan Zeng, Zhifei He, Zherui Tan, Renliang Geng, Menghua Liu, Linglin Xia, Deyu Lin
Summary: The study aimed to provide a reasonable iBeacon networking deployment scheme to enhance positioning accuracy by exploring signal propagation characteristics and path loss. By collecting and filtering signal samples, as well as using propagation models to simulate iBeacon signal propagation, the best applicable deployment scenario was determined based on regression analysis of sample data under different scenarios.
INTERNATIONAL JOURNAL OF SENSOR NETWORKS
(2021)
