Article
Computer Science, Artificial Intelligence
Meng Hu, Yanting Guo, Degang Chen, Eric C. C. Tsang, Qingshuo Zhang
Summary: The construction of fuzzy relations is crucial in fuzzy rough sets, and relations generated by soft distances are more robust. Two enhanced fuzzy similarity relations are proposed to improve attribute reduction and classification, using neighborhood and decision information. The algorithm is validated using gene expression profiles and demonstrates strong noise resistance and selection of tumor-related genes.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Information Systems
Zhong Yuan, Hongmei Chen, Tianrui Li, Zeng Yu, Binbin Sang, Chuan Luo
Summary: The study proposed a generalized unsupervised mixed attribute reduction model based on fuzzy rough sets and designed a specific algorithm FRUAR. Experimental results showed that the FRUAR algorithm can select fewer attributes to maintain or improve the performance of learning algorithms.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Information Systems
Jianhua Dai, Xiongtao Zou, Wei-Zhi Wu
Summary: In this study, a new framework of fuzzy fl-covering rough set models is proposed, and the existing methods are improved by constructing new fuzzy fl-neighborhood operators and defining the fuzzy fl-covering relation. Experimental results demonstrate the significant advantages of the proposed method.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Xia Ji, JianHua Peng, Peng Zhao, Sheng Yao
Summary: Attribute reduction is a crucial step in data analysis, and the attribute reduction method based on neighborhood rough sets (NRS) is commonly used. However, this method has high time complexity due to its reliance on grid search for radius selection. Granular ball neighborhood rough sets (GBNRS) offer more generality and flexibility than NRS, but they may delete granular balls at class boundaries, leading to the loss of class boundary information. To address these issues, a fuzzy granular ball is defined, and an extended rough sets model (FGBERS) is proposed based on this ball. Experimental results show that FGBERS outperforms GBNRS in terms of classification accuracy, especially on high-dimensional datasets.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Theory & Methods
Xiaoya Che, Degang Chen, Jusheng Mi
Summary: This study proposes a method based on fuzzy rough sets for local attribute reduction to calculate and define the crucial features for each local label class, constructing local and global label correlations, and achieving the transformation of labels into independent subsets.
FUZZY SETS AND SYSTEMS
(2022)
Article
Computer Science, Information Systems
Zhouming Ma, Jusheng Mi, Yiting Lin, Jinjin Li
Summary: Variable precision rough set (VPRS) has been widely studied as an essential way of knowledge representation and acquisition in uncertainty theory. This paper investigates the corresponding CVPRS model based on a covering-based rough set model, and systematically studies its algebraic structures and properties. An attribute reduction approach is proposed for a covering-based decision information system using the CVPRS model, and the performances of different boundary operators and related indices in these reduction methods are compared. Necessity rules and possibility rules extraction methods corresponding to decision classes are established, and their validity and security are theoretically verified.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Lei Ma, Chuan Luo, Tianrui Li, Hongmei Chen, Dun Liu
Summary: With the accumulation of interesting data in various application fields, incremental datasets are becoming more common. However, selecting informative attributes from dynamically changing datasets poses challenges. Therefore, an incremental processing mechanism is desired to update the attribute reducts efficiently. In this paper, a novel dynamic graph-based fuzzy rough attribute reduction approach is proposed to handle the maintenance of fuzzy rough attribute reduction in dynamic data, which outperforms existing methods in terms of speed and quality preservation.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2023)
Article
Computer Science, Information Systems
Changzhong Wang, Yang Huang, Weiping Ding, Zehong Cao
Summary: Fuzzy rough sets combined with the concept of self-information are used to construct four uncertainty measures to evaluate the classification ability of attribute subsets. The fourth measure, relative decision self-information, is proven to be better for attribute reduction. A greedy algorithm is designed for attribute reduction, and the effectiveness of the method is validated through experimental results.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Xianyong Zhang, Yunrui Fan, Yuesong Yao, Jilin Yang
Summary: This paper proposes class-specific attribute reducts based on neighborhood rough sets to achieve optimal identification and robust processing of specific classes. The neighborhood class-specific reducts extend the existing class-specific reducts and provide a hierarchical mechanism for the neighborhood classification-based reducts, facilitating wide applications of class-pattern processing.
JOURNAL OF INTELLIGENT & FUZZY SYSTEMS
(2022)
Article
Computer Science, Information Systems
Xianyong Zhang, Jiefang Jiang
Summary: This study improves the Variable Precision Multigranulation Fuzzy Rough Sets (VP-MFRSs) by proposing Decision-Theoretic Multigranulation Fuzzy Rough Sets (DT-MFRSs) which systematically fuse the multigranulation maximum and minimum. DT-MFRSs provide tri-level analysis of measurement, modeling, and reduction via three-way decisions. The study extends and improves VP-MFRSs by introducing optimistic, pessimistic, and compromised models, and enhances uncertainty optimization through a new reduction criteria.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Meng Hu, Eric C. C. Tsang, Yanting Guo, Degang Chen, Weihua Xu
Summary: This study introduces a novel attribute reduction method based on weighted neighborhood relations, which fully mines the correlation between attributes and decisions, assigning higher weights to attributes with higher correlation, achieving good performance results.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Pandu Sowkuntla, P. S. V. S. Sai Prasad
Summary: Fuzzy-rough set theory is an efficient method for attribute reduction, but current approaches are inefficient for large data sets. This paper introduces a fuzzy discernibility matrix-based attribute reduction accelerator that performs better than existing methods in terms of computational efficiency.
APPLIED INTELLIGENCE
(2022)
Article
Computer Science, Information Systems
Guoqiang Wang, Tianrui Li, Pengfei Zhang, Qianqian Huang, Hongmei Chen
Summary: Local rough set models are effective for handling large data sets with small amounts of labeled data, improving computational performance significantly. The double-local rough set framework introduces the concept of local equivalence classes and defines lower deletion matrix, upper addition matrix, and upper deletion matrix. Proposed algorithms in double-local rough sets outperform original counterparts in attribute reduction and knowledge discovery.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Interdisciplinary Applications
Jin Ye, Jianming Zhan, Zeshui Xu
Summary: This paper proposes a novel decision-making method based on fuzzy rough sets to transform uncertain data into intuitionistic fuzzy data, establish a new MADM method, and introduce intuitionistic fuzzy weights and global intuitionistic fuzzy thresholds.
COMPUTERS & INDUSTRIAL ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Jiali He, Liangdong Qu, Zhihong Wang, Yiying Chen, Damei Luo, Ching-Feng Wen
Summary: This paper investigates attribute reduction in an incomplete categorical decision information system (ICDIS) based on fuzzy rough sets. An attribute reduction algorithm is proposed and experiments show that it outperforms existing algorithms.
ARTIFICIAL INTELLIGENCE REVIEW
(2022)
Article
Computer Science, Artificial Intelligence
Ming Yang, Hongguang Ma, Xiang Li, Changjing Shang, Qiang Shen
Summary: This article studies the bus bridging problem in public transportation systems, where passenger demand is represented as parametric interval-valued fuzzy variables and their associated uncertainty distribution sets. A distributionally robust fuzzy optimization model is proposed to minimize the maximum travel time and find the optimal scheme for vehicle allocation, route selection, and frequency determination. The proposed approach is verified using real-world uncertain parameters and validated to provide a better uncertainty-immunized solution.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Li -Jiang Li, Sheng-Lin Zhou, Fei Chao, Xiang Chang, Longzhi Yang, Xiao Yu, Changjing Shang, Qiang Shen
Summary: This paper presents a network compression method using knowledge distillation technology to develop concise neural network-based controllers that strike a balance between control performance and computational costs. The method involves training a full-size teacher model, pruning it to obtain a compact network, and then further training the compact network as a student model using knowledge distillation. Experimental results demonstrate that the student models with fewer neurons achieve similar performance to the teacher models for intelligent dynamic control but with faster convergence speed.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Jing Yang, Chanyue Wu, Tengfei You, Dong Wang, Ying Li, Changjing Shang, Qiang Shen
Summary: In this study, a new dual pyramid model is proposed for hyperspectral image super resolution. It utilizes a novel hierarchical spatial and spectral fusion method to progressively estimate the high resolution image. Qualitative and quantitative evaluations demonstrate its outstanding performance over large scale factors compared to other spatio-spectral fusion based super resolution techniques.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Xiaowei Gu, Miqing Li, Liang Shen, Guolin Tang, Qiang Ni, Taoxin Peng, Qiang Shen
Summary: Zero-order intelligent systems have shown strong performance in data stream classification with high model transparency. However, the lack of optimality in identified prototypes hinders the classification performance. To address this, a novel multiobjective optimization approach is proposed in this article, combining training error minimization and intracluster variance minimization. Experimental studies demonstrate the effectiveness of the approach in improving the classification performance of zero-order intelligent systems.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Mou Zhou, Changjing Shang, Guobin Li, Liang Shen, Nitin Naik, Shangzhu Jin, Jun Peng, Qiang Shen
Summary: Fuzzy rule interpolation (FRI) is enhanced by using a novel transformation-based approach that utilizes the Mahalanobis distance metric for rule selection. By transforming the rule base into a coordinate system, instances of the same category are brought closer and instances of different categories are moved further apart. This allows for more accurate selection of neighboring rules for interpolation when a new observation matches no rules.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Computer Science, Information Systems
Jiagang Liu, Ju Ren, Yongmin Zhang, Xuhong Peng, Yaoxue Zhang, Yuanyuan Yang
Summary: In this paper, a dependent task offloading framework (COFE) is proposed, which allows mobile devices to offload compute-intensive tasks with dependent constraints to the MEC-Cloud system to improve user experience. The task offloading problem is formulated as an average makespan minimization problem, and a heuristic ranking-based algorithm is proposed to assign the offloaded tasks. Theoretical analysis proves the stability of the system under the proposed algorithm, and extensive simulations validate its effectiveness in reducing average makespan and deadline violation probabilities.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Xuan Hou, Yunpeng Bai, Yefan Xie, Huibin Ge, Ying Li, Changjing Shang, Qiang Shen
Summary: Deep learning has shown great potential in SAR imagery change detection, but it requires a large amount of labeled samples, which is a tedious and time-consuming task. In addition, sample imbalance is a challenge for existing change detection techniques. To address these problems, a Deep Collaborative semi-supervised learning Framework with Class-Rebalancing (DCF-CRe) is proposed for SAR imagery change detection, by using Convolutional Neural Network (CNN) and deep clustering.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Automation & Control Systems
Jialin Liu, Fei Chao, Longzhi Yang, Chih-Min Lin, Changjing Shang, Qiang Shen
Summary: This article presents a novel metalearning method that controls the gradient descent process in a neural network by limiting the model parameters in a low-dimensional latent space. It also introduces an alternative design of the decoder with shared weights to reduce the number of parameters. Experimental results show that the proposed approach outperforms the state of the art in classification tasks.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Computer Science, Artificial Intelligence
Guanli Yue, Yanpeng Qu, Longzhi Yang, Changjing Shang, Ansheng Deng, Fei Chao, Qiang Shen
Summary: Fuzzy clustering is a method that uses partial memberships to decompose data into clusters, and it demonstrates comparable performance in knowledge exploitation when dealing with incomplete information. This article proposes a new fuzzy-rough intrigued harmonic discrepancy clustering (HDC) algorithm that effectively handles complex data distribution and improves clustering performance.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Dong Wang, Chanyue Wu, Yunpeng Bai, Ying Li, Changjing Shang, Qiang Shen
Summary: This paper proposes a multitask network (MTNet) to achieve joint multispectral (MS) pansharpening for images acquired by different satellites. The MTNet shares generic knowledge between datasets via a task-agnostic subnetwork (TASNet) and adapts this knowledge to specific satellites using task-specific subnetworks (TSSNets). It also introduces band-aware dynamic convolutions (BDConvs) to accommodate various ground scenes and bands. Experimental results demonstrate that the proposed approach outperforms existing state-of-the-art (SOTA) techniques across different datasets.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Mizi Han, Yanpeng Qu, Neil MacParthalain, Changjing Shang, Zihan Yao, Qiang Shen
Summary: Three-valued reasoning or three-way decision modelling theory (3WD) is a natural and intuitive approach to handling uncertainty. This paper proposes a linear reconstruction neighbourhood membership and representation-based decision-theoretic rough set (RDTRS) approach to improve 3WD classification models. The experimental results demonstrate improved classification performance for both benchmark and face image datasets.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2023)
Article
Automation & Control Systems
Mou Zhou, Changjing Shang, Qiang Shen
Summary: Fuzzy rule interpolation empowers fuzzy rule-based systems to infer even with sparse rule bases. This article introduces a groundbreaking rule-ranking-based method, RT-FRI, which streamlines the rule selection procedure by using ranking scores produced through aggregation functions. Experimental results show that RT-FRI is highly efficient and accurate.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Computer Science, Artificial Intelligence
Qingping Zheng, Ling Zheng, Jiankang Deng, Ying Li, Changjing Shang, Qiang Shen
Summary: In this paper, a Transformer-based Hierarchical Dynamic Decoder (T-HDDNet) is proposed for salient object detection. The method utilizes a self-attention mechanism to extract features and has a powerful capability of learning global cues. With a dynamic dual upsampling mechanism and a dynamic feature fusion unit, it achieves accurate saliency maps of high resolution in a data-driven manner.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Automation & Control Systems
Ruiqi Wu, Fei Chao, Changle Zhou, Xiang Chang, Longzhi Yang, Changjing Shang, Zihao Zhang, Qiang Shen
Summary: This article proposes a robotic writing framework based on a robotic hand-eye coordination method. By constructing a vision-motor network and a motor-vision network, the proposed method successfully writes strokes of Chinese characters. The underlying research of this method can be applied to other areas, such as human-robot motion mimicking.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Automation & Control Systems
Quanfeng Li, Zhihua Guo, Fei Chao, Xiang Chang, Longzhi Yang, Chih-Min Lin, Changjing Shang, Qiang Shen
Summary: This article introduces a new robot calligraphy system that can learn writing sequences with limited information, utilizing a GRU network and prelabeled trajectory sequence vector for optimal writing results. The proposed evaluation method considers shape, trajectory sequence, and structural information to ensure writing quality. Experimental results show the system's competitive writing performance and diverse outcomes.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)