Article
Computer Science, Artificial Intelligence
Guanxiang Hu, Wei He, Chao Sun, Hailong Zhu, Kangle Li, Li Jiang
Summary: Classification tasks are important in machine learning, but the problem of class imbalance can significantly affect classifier performance. This paper proposes a hierarchical belief rule-based system that integrates expert knowledge and utilizes extreme gradient boosting for feature selection to address class imbalance. By transforming multi-classification problems into binary classification problems and making precise predictions, class imbalance is alleviated.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Abhishek Dixit, Ashish Mani
Summary: Class imbalance learning is an important branch of machine learning, where the efficiency of classifiers is affected due to imbalanced datasets. The Synthetic Minority Oversampling Technique (SMOTE) is the most successful solution for data imbalance, but it faces challenges from noise and borderline examples. To address these issues, a novel oversampling filter-based method called SMOTE-TLNN-DEPSO is proposed, which generates synthetic samples using SMOTE, applies error detection through a two-layer natural neighbors' technique, and optimizes noisy samples using a hybrid variant of the DEPSO algorithm.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Theory & Methods
Mircea Cimpoeas, Adrian Gabriel Neacsu
Summary: This article discusses the relationship between covering and partition in fuzzy sets. It geometrically constructs an isomorphism between both categories, which can be used to derive bijections between fuzzy partitions and coverings with a finite number of fuzzy sets. Additionally, an isomorphism is established between the category of coverings with n fuzzy sets and a subcategory of partition with n fuzzy sets satisfying certain conditions.
FUZZY SETS AND SYSTEMS
(2023)
Article
Computer Science, Information Systems
Lin Sun, Mengmeng Li, Weiping Ding, En Zhang, Xiaoxia Mu, Jiucheng Xu
Summary: This paper proposes a novel adaptive fuzzy neighborhood-based feature selection method for imbalanced data with adaptive synthetic over-sampling. It addresses the limitations of manually setting fuzzy neighborhood radius and potential ignorance of boundary regions, and achieves effective classification results.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Theory & Methods
Jie Liu
Summary: This paper extends the FSVM-CIL method proposed by Rukshan Batuwita and Vasile Palade, introduces a new Gaussian fuzzy function, and validates the effectiveness of the proposed method on imbalanced data through experiments.
FUZZY SETS AND SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Wanli Huang, Yanhong She, Xiaoli He, Weiping Ding
Summary: This article presents an incremental feature selection approach for hierarchical classification in the era of big data. By employing the sibling strategy, theoretical analysis, and algorithmic design, two incremental algorithms are proposed and demonstrated to be effective and feasible.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Mathematics
Jinghong Zhang, Yingying Li, Bowen Liu, Hao Chen, Jie Zhou, Hualong Yu, Bin Qin
Summary: With the expansion of data scale and diversity, the issue of class imbalance has become increasingly salient. To address these challenges, a novel fuzzy classifier is proposed that can handle classification tasks with class-imbalanced data.
Article
Computer Science, Artificial Intelligence
Fei Gao
Summary: Sparse rule base is a common problem in fuzzy rule-based systems. This paper presents a density-based fuzzy rule interpolation method that adaptively selects the closest rules with high similarity to the inputs. The method has been verified through fifteen classification benchmarks, demonstrating its effectiveness and efficiency.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Maria Arteaga, Maria Jose Gacto, Marta Galende, Jesus Alcala-Fdez, Rafael Alcala
Summary: Research on imbalance is mainly focused on classification, with little attention to regression. This paper proposes two evolutionary algorithms based on fuzzy rules to actively address imbalance in regression problems and improve algorithm performance. Experimental results show that these methods outperform previous approaches.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Xin Yang, Yujie Li, Dun Liu, Tianrui Li
Summary: The approximation learning of fuzzy concepts associated with fuzzy rough sets and three-way decisions is an important technology for handling uncertain knowledge. This article explores the connection and interplay between fuzzy rough approximations and three-way approximations, proposing a hierarchical fuzzy rough approximation method in a dynamic fuzzy open-world environment. The article also discusses the interpretation and representation of fuzzy three-way regions in fuzzy rough sets. Experimental results demonstrate the effectiveness of the proposed hierarchical approximation learning models.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Jaeyeon Jang, Chang Ouk Kim
Summary: Fault detection is crucial in industrial processes, but the imbalanced and high-dimensional datasets pose challenges. This paper proposes a novel model, UB-SOM, to address class imbalance and high-dimensionality problems. Experimental results demonstrate that UB-SOM can significantly improve fault detection performance through data preprocessing.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Junnan Li, Qingsheng Zhu, Quanwang Wu, Zhiyong Zhang, Yanlu Gong, Ziqing He, Fan Zhu
Summary: The novel filtering-based oversampling method SMOTE-NaN-DE proposed in this paper effectively addresses the noise problem in SMOTE-based methods and improves the decision boundary by optimizing the examples detected by error detection techniques using differential evolution. It is more suitable for datasets with more noise, especially class noise, and has been validated through comparison experiments on artificial and real datasets.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Information Systems
J. Sanz, M. Sesma-Sara, H. Bustince
Summary: This paper introduces a new fuzzy association rule-based classifier, FARCI, for directly tackling imbalanced classification problems. Experimental results show the superiority of the new method in terms of performance, F-score, and rule base size when compared to other algorithm modification approaches.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Gabriella Casalino, Giovanna Castellano, Ciro Castiello, Corrado Mencar
Summary: Fuzziness has an impact on the behavior of Fuzzy Rule-Based Classifiers, but only in regions where the classification confidence is low. Therefore, in Explainable Artificial Intelligence, fuzziness is beneficial in FRBCs only when accompanied by an explanation of the output confidence.
Article
Computer Science, Artificial Intelligence
Saeed Zeraatkar, Fatemeh Afsari
Summary: Imbalanced data classification is a complex issue where traditional classifiers perform poorly due to imbalanced class distribution. This paper proposes a solution involving resampling the data in two phases, oversampling and undersampling, using robust extensions of KNN classifiers based on interval-valued fuzzy and intuitionistic fuzzy sets. The proposed method shows potential in handling both synthetic and real-world data sets with different levels of noise and borderlines.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Francisco J. Pulgar, Francisco Charte, Antonio J. Rivera, Maria J. del Jesus
INFORMATION FUSION
(2020)
Article
Genetics & Heredity
Nadia Pinto, Vania Pereira, Carmen Tomas, Silvia Loiola, Elizeu F. Carvalho, Nidia Modesti, Mariana Maxzud, Valeria Marcucci, Hortensia Cano, Regina Cicarelli, Bianca Januario, Ana Bento, Pedro Brito, German Burgos, Elius Paz-Cruz, Laura Diez-Juarez, Silvia Vannelli, Maria de Lurdes Pontes, Gabriela Berardi, Sandra Furfuro, Alberto Fernandez, Denilce Sumita, Cecilia Bobillo, Maria Gabriela Garcia, Leonor Gusmao
FORENSIC SCIENCE INTERNATIONAL-GENETICS
(2020)
Article
Computer Science, Information Systems
Francisco J. Pulgar, Francisco Charte, Antonio J. Rivera, Maria J. del Jesus
Summary: The paper introduces a new classifier, ClEnDAE, which uses ensemble methods and DAE to reduce dimensionality of input space and improve predictive performance. Experimental results show that the algorithm outperforms other traditional methods in classification.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Jose Daniel Pascual-Triana, David Charte, Marta Andres Arroyo, Alberto Fernandez, Francisco Herrera
Summary: Data Science and Machine Learning play crucial roles for companies and research institutions, with supervised classification allowing for class prediction of new samples but certain properties may make datasets challenging to classify. Data complexity metrics are extensively used to evaluate datasets, providing information on intrinsic data characteristics to assess classifier compatibility. However, most metrics focus on a single aspect of the data, potentially inadequate for comprehensive dataset evaluation. This research revisits complexity metrics based on data morphology, proposing a new family of metrics named Overlap Number of Balls which aim to provide better estimates for class overlap and correlate with classification performance.
KNOWLEDGE AND INFORMATION SYSTEMS
(2021)
Article
Computer Science, Information Systems
Maria Jose Basgall, Marcelo Naiouf, Alberto Fernandez
Summary: FDR2-BD presents a methodological data condensation approach for reducing tabular big datasets in classification problems, which combines feature selection and uniform sampling reduction to maintain predictive quality within a user-defined threshold. The method shows robustness and scalability, outperforming existing solutions in reduction percentages while maintaining representativeness of the original data information.
Article
Computer Science, Artificial Intelligence
J. A. Fdez-Sanchez, J. D. Pascual-Triana, A. Fernandez, F. Herrera
Summary: The study introduces a novel method TC-ND based on hierarchical decompositions for obtaining interpretable multi-class models. This method creates a binary-based hierarchical class structure and discards meta-classes based on confidence levels, achieving a modular and easily understandable decomposition of multi-class problems.
Article
Automation & Control Systems
Miriam Seoane Santos, Pedro Henriques Abreu, Alberto Fernandez, Julian Luengo, Joao Santos
Summary: This work examines the impact of distance functions on K-Nearest Neighbours imputation of incomplete datasets. The experiments show that distance computation is significantly affected by missing data, and provide guidelines for selecting appropriate distance functions based on data characteristics and research objectives.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Fatemeh Aghaeipoor, Mohammad Masoud Javidi, Alberto Fernandez
Summary: This article introduces an interpretable fuzzy classifier for Big Data, aiming to boost explainability by learning a compact yet accurate fuzzy model. Developed in a cell-based distributed framework, IFC-BD goes through three working stages of initial rule learning, rule generalization, and heuristic rule selection to move from a high number of specific rules to fewer, more general and confident rules. The proposed algorithm was found to improve the explainability and predictive performance of fuzzy rule-based classifiers in comparison to state-of-the-art approaches.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Sebastian Maldonado, Carla Vairetti, Alberto Fernandez, Francisco Herrera
Summary: This paper introduces a novel SMOTE variant based on the weighted Minkowski distance, which is used to address the class-imbalance problem in high-dimensional settings. By prioritizing features that are more relevant for the classification task, the proposed method provides a better definition of the neighborhood. Furthermore, it offers the additional advantage of feature selection and performs well in handling class overlap and hubness issues.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
A. M. Garcia-Vico, C. J. Carmona, P. Gonzalez, M. J. del Jesus
Summary: This paper presents a distributed method based on evolutionary fuzzy systems for extracting and fusing patterns from data streams, and analyzes the adaptability and quality of the proposed method.
INFORMATION FUSION
(2023)
Article
Pharmacology & Pharmacy
Carlos Pedraz-Valdunciel, Stavros Giannoukakos, Ana Gimenez-Capitan, Diogo Fortunato, Martyna Filipska, Jordi Bertran-Alamillo, Jillian W. P. Bracht, Ana Drozdowskyj, Joselyn Valarezo, Natasa Zarovni, Alberto Fernandez-Hilario, Michael Hackenberg, Andres Aguilar-Hernandez, Miguel Angel Molina-Vila, Rafael Rosell
Summary: This study validates the use of the nCounter platform for multiplexed EV-circRNA expression studies in LC patient samples and successfully develops a prognostic signature for LC.
Article
Engineering, Biomedical
Jose P. Amorim, Pedro H. Abreu, Alberto Fernandez, Mauricio Reyes, Joao Santos, Miguel H. Abreu
Summary: Healthcare agents are collecting large amounts of patient data, particularly in oncology. Decision-support systems based on deep learning techniques have been approved for clinical use, but their interpretability remains a barrier to their widespread adoption. This article aims to provide oncologists with a guide on how these methods make decisions and strategies to explain them. Theoretical concepts were illustrated using oncological examples and a literature review was conducted to identify research works in the field. The majority of studies focused on explaining the importance of tumor characteristics in predictions using multilayer perceptrons and convolutional neural networks. However, achieving interpretability while maintaining performance remains a significant challenge for artificial intelligence.
IEEE REVIEWS IN BIOMEDICAL ENGINEERING
(2023)
Article
Biochemistry & Molecular Biology
Silvia D'Ambrosi, Stavros Giannoukakos, Mafalda Antunes-Ferreira, Carlos Pedraz-Valdunciel, Jillian W. P. Bracht, Nicolas Potie, Ana Gimenez-Capitan, Michael Hackenberg, Alberto Fernandez Hilario, Miguel A. Molina-Vila, Rafael Rosell, Thomas Wurdinger, Danijela Koppers-Lalic
Summary: This study investigates the synergistic contribution of circRNA and mRNA derived from blood platelets as biomarkers for lung cancer detection. Using a comprehensive bioinformatics pipeline, platelet-derived circRNA and mRNA from non-cancer individuals and lung cancer patients were analyzed. Machine learning algorithms were used to generate predictive classification models based on an optimal selected signature. The study demonstrates the potential of a multi-analyte-based approach using platelet-derived biomarkers for lung cancer detection.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Fatemeh Aghaeipoor, Mohammad Sabokrou, Alberto Fernandez
Summary: The explainability of deep neural networks has become a topic of interest, and this article proposes a fuzzy rule-based explainer system that helps understand the functioning of these networks. These systems maintain accuracy while reducing complexity and improving comprehensibility.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Clinical Neurology
Marc Norman, Sarah J. Wilson, Sallie Baxendale, William Barr, Cady Block, Robyn M. Busch, Alberto Fernandez, Erik Hessen, David W. Loring, Carrie R. McDonald, Bruce P. Hermann
Summary: This paper addresses the lack of an international diagnostic taxonomy for cognitive disorders in epilepsy and proposes a consensus-based classification system framework as a solution.
Article
Computer Science, Artificial Intelligence
Hao Yang, Min Wang, Zhengfei Yu, Hang Zhang, Jinshen Jiang, Yun Zhou
Summary: In this paper, a novel method called CSTTA is proposed for test time adaptation (TTA), which utilizes confidence-based optimization and sample reweighting to better utilize sample information. Extensive experiments demonstrate the effectiveness of the proposed method.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Jin Liu, Ju-Sheng Mi, Dong-Yun Niu
Summary: This article focuses on a novel method for generating a canonical basis for decision implications based on object-induced operators (OE operators). The logic of decision implication based on OE operators is described, and a method for obtaining the canonical basis for decision implications is given. The completeness, nonredundancy, and optimality of the canonical basis are proven. Additionally, a method for generating true premises based on OE operators is proposed.
KNOWLEDGE-BASED SYSTEMS
(2024)
Review
Computer Science, Artificial Intelligence
Kun Bu, Yuanchao Liu, Xiaolong Ju
Summary: This paper discusses the importance of sentiment analysis and pre-trained models in natural language processing, and explores the application of prompt learning. The research shows that prompt learning is more suitable for sentiment analysis tasks and can achieve good performance.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Xiangjun Cai, Dagang Li
Summary: This paper presents a new decomposition mechanism based on learned decomposition mapping. By using a neural network to learn the relationship between original time series and decomposed results, the repetitive computation overhead during rolling decomposition is relieved. Additionally, extended mapping and partial decomposition methods are proposed to alleviate boundary effects on prediction performance. Comparative studies demonstrate that the proposed method outperforms existing RDEMs in terms of operation speed and prediction accuracy.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Xu Wu, Yang Liu, Jie Tian, Yuanpeng Li
Summary: This paper proposes a blockchain-based privacy-preserving trust management architecture, which adopts federated learning to train task-specific trust models and utilizes differential privacy to protect device privacy. In addition, a game theory-based incentive mechanism and a parallel consensus protocol are proposed to improve the accuracy of trust computing and the efficiency of consensus.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Zaiyang Yu, Prayag Tiwari, Luyang Hou, Lusi Li, Weijun Li, Limin Jiang, Xin Ning
Summary: This study introduces a 3D view-based approach that effectively handles occlusions and leverages the geometric information of 3D objects. The proposed method achieves state-of-the-art results on occluded ReID tasks and exhibits competitive performance on holistic ReID tasks.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Yongliang Shi, Runyi Yang, Zirui Wu, Pengfei Li, Caiyun Liu, Hao Zhao, Guyue Zhou
Summary: Neural implicit representations have gained attention due to their expressive, continuous, and compact properties. However, there is still a lack of research on city-scale continual implicit dense mapping based on sparse LiDAR input. In this study, a city-scale continual neural mapping system with a panoptic representation is developed, incorporating environment-level and instance-level modeling. A tailored three-layer sampling strategy and category-specific prior are proposed to address the challenges of representing geometric information in city-scale space and achieving high fidelity mapping of instances under incomplete observation.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Ruihan Hu, Zhi-Ri Tang, Rui Yang, Zhongjie Wang
Summary: Mesh data is crucial for 3D computer vision applications worldwide, but traditional deep learning frameworks have struggled with handling meshes. This paper proposes MDSSN, a simple mesh computation framework that models triangle meshes and represents their shape using face-based and edge-based Riemannian graphs. The framework incorporates end-to-end operators inspired by traditional deep learning frameworks, and includes dedicated modules for addressing challenges in mesh classification and segmentation tasks. Experimental results demonstrate that MDSSN outperforms other state-of-the-art approaches.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Buliao Huang, Yunhui Zhu, Muhammad Usman, Huanhuan Chen
Summary: This paper proposes a novel semi-supervised conditional normalizing flow (SSCFlow) algorithm that combines unsupervised imputation and supervised classification. By estimating the conditional distribution of incomplete instances, SSCFlow facilitates imputation and classification simultaneously, addressing the issue of separated tasks ignoring data distribution and label information in traditional methods.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Deeksha Varshney, Asif Ekbal, Erik Cambria
Summary: This paper focuses on the neural-based interactive dialogue system that aims to engage and retain humans in long-lasting conversations. It proposes a new neural generative model that combines step-wise co-attention, self-attention-based transformer network, and an emotion classifier to control emotion and knowledge transfer during response generation. The results from quantitative, qualitative, and human evaluation show that the proposed models can generate natural and coherent sentences, capturing essential facts with significant improvement over emotional content.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Junchen Ye, Weimiao Li, Zhixin Zhang, Tongyu Zhu, Leilei Sun, Bowen Du
Summary: Modeling multivariate time series has long been a topic of interest for scholars in various fields. This paper introduces MvTS, an open library based on Pytorch, which provides a unified framework for implementing and evaluating these models. Extensive experiments on public datasets demonstrate the effectiveness and universality of the models reproduced by MvTS.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Reham R. Mostafa, Ahmed M. Khedr, Zaher Al Aghbari, Imad Afyouni, Ibrahim Kamel, Naveed Ahmed
Summary: Feature selection is crucial in classification procedures, but it faces challenges in high-dimensional datasets. To overcome these challenges, this study proposes an Adaptive Hybrid-Mutated Differential Evolution method that incorporates the mechanics of the Spider Wasp Optimization algorithm and the concept of Enhanced Solution Quality. Experimental results demonstrate the effectiveness of the method in terms of accuracy and convergence speed, and it outperforms contemporary cutting-edge algorithms.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Ti Xiang, Pin Lv, Liguo Sun, Yipu Yang, Jiuwu Hao
Summary: This paper introduces a Track Classification Model (TCM) based on marine radar, which can effectively recognize and classify shipping tracks. By using a feature extraction network with multi-feature fusion and a dataset production method to address missing labels, the classification accuracy is improved, resulting in successful engineering application in real scenarios.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Zhihao Zhang, Yuan Zuo, Chenghua Lin, Junjie Wu
Summary: This paper proposes a novel unsupervised context-aware quality phrase mining framework called LMPhrase, which is built upon large pre-trained language models. The framework mines quality phrases as silver labels using a parameter-free probing technique on the pre-trained language model BERT, and formalizes the phrase tagging task as a sequence generation problem by fine-tuning on the Sequence to-Sequence pre-trained language model BART. The results of extensive experiments show that LMPhrase consistently outperforms existing competitors in two different granularity phrase mining tasks.
KNOWLEDGE-BASED SYSTEMS
(2024)
Article
Computer Science, Artificial Intelligence
Kemal Buyukkaya, M. Ozan Karsavuran, Cevdet Aykanat
Summary: The study aims to investigate the hybrid parallelization of the Stochastic Gradient Descent (SGD) algorithm for solving the matrix completion problem on a high-performance computing platform. A hybrid parallel decentralized SGD framework with asynchronous inter-process communication and a novel flexible partitioning scheme is proposed to achieve scalability up to hundreds of processors. Experimental results on real-world benchmark datasets show that the proposed algorithm achieves 6x higher throughput on sparse datasets compared to the state-of-the-art, while achieving comparable throughput on relatively dense datasets.
KNOWLEDGE-BASED SYSTEMS
(2024)