Article
Computer Science, Information Systems
Tapas Bhadra, Sanghamitra Bandyopadhyay
Summary: The paper proposes a novel supervised feature selection approach based on dense subgraph discovery. The algorithm proceeds in two phases to select features with maximal average class relevance, minimal average pairwise redundancy, and good discriminating power. Experimental results show the proposed approach is competitive with conventional and state-of-the-art algorithms in supervised feature selection.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Emrah Hancer
Summary: Fuzzy mutual information is a popular method in information theory for quantifying the information between random variables, capable of handling different types of variables effectively. Recently, it has been integrated into evolutionary filter feature selection approaches to significantly improve the computational efficiency and performance of classification algorithms on real-world datasets.
Article
Biochemical Research Methods
Tapas Bhadra, Saurav Mallik, Amir Sohel, Zhongming Zhao
Summary: This article proposes a novel unsupervised feature selection method by combining hierarchical feature clustering with singular value decomposition. The experimental results demonstrate that the proposed algorithm performs well against several state-of-the-art methods of feature selection in terms of various evaluation criteria. The analysis of cancer genomics data identifies a candidate gene-marker, EREG, which is important for biomarker discovery in precision medicine.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Article
Computer Science, Artificial Intelligence
Francisco Macedo, Rui Valadas, Eunice Carrasquinha, M. Rosario Oliveira, Antonio Pacheco
Summary: Feature selection is an essential preprocessing technique to improve regression and classification tasks by reducing dimensionality. DMIM method, based on Decomposed Mutual Information Maximization, overcomes complementarity penalization issue in existing methods, showing better classification performance in experiments.
Article
Computer Science, Artificial Intelligence
Pawel Teisseyre, Jaesung Lee
Summary: In this paper, a multilabel all-relevant feature selection task is discussed for multi-label classification. The all-relevant methods aim to identify all features related to target labels, including weakly relevant features. The paper proposes a relevancy score calculation method based on conditional mutual information and a testing procedure for separating relevant and irrelevant features.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Tapas Bhadra, Ujjwal Maulik
Summary: In this article, an unsupervised feature selection algorithm based on the SEA algorithm is proposed. The algorithm identifies dense subgraphs within a weighted graph and presents the feature selection problem as obtaining dense sub-feature spaces. It consists of two stages, finding dense feature subgraphs and identifying representative features. The algorithm does not require the user to provide the number of features to be selected and achieves high accuracy scores.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Peng Zhou, Ni Wang, Shu Zhao
Summary: This paper focuses on the interaction of features within and between streaming groups, proposing an Online Group Streaming Feature Selection method named OGSFS-FI, which consists of two stages: online intra-group selection and online inter-group selection. The method utilizes a new pair selection strategy and the elastic net method for efficient and effective feature selection.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Zhaolong Ling, Ying Li, Yiwen Zhang, Kui Yu, Peng Zhou, Bo Li, Xindong Wu
Summary: Causal feature selection has received increasing attention. However, existing algorithms have high computational complexity. To address this, this paper proposes a novel algorithm called CFS-MI, which analyzes the unique performance of causal features in mutual information and reduces computational complexity by separating pairwise comparisons in two stages. Experimental results demonstrate that CFS-MI achieves comparable accuracy and superior computational efficiency compared to 7 state-of-the-art algorithms.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Ahmad Esfandiari, Hamid Khaloozadeh, Faezeh Farivar
Summary: This paper introduces a multivariate filter feature selection method called interaction-based feature clustering (IFC), which is cost-effective in terms of computational cost while achieving high classification accuracy. The proposed method ranks features based on the symmetric uncertainty criterion and performs feature clustering by calculating their interactive weight as a similarity measure. Experimental results show that the IFC algorithm is more efficient than comparable methods in terms of classification accuracy and computational time.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2023)
Article
Computer Science, Artificial Intelligence
Tingting Pan, Witold Pedrycz, Jie Yang, Wei Wu, Yulin Zhang
Summary: In this paper, a new iterative ensemble classifier (C-ILEO) is proposed for imbalanced data. The iterative learning process and ensemble operating process are used to improve classification performance by selecting a small number of features and optimizing class weights. Experimental results show that C-ILEO outperforms other algorithms and methods on imbalanced datasets.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Pei Huang, Xiaowei Yang
Summary: Unsupervised feature selection is an important topic in the fields of machine learning, pattern recognition and data mining. A novel method called AGDS is proposed to address the issues in feature selection by utilizing adaptive graph and dependency score.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
Snehalika Lall, Debajyoti Sinha, Abhik Ghosh, Debarka Sengupta, Sanghamitra Bandyopadhyay
Summary: The study introduces a feature selection algorithm based on copula that maximizes feature relevance and minimizes redundant information. The proposed CBFS algorithm competes well in maximizing classification accuracy on real and synthetic datasets and demonstrates better noise tolerance compared to other methods.
PATTERN RECOGNITION
(2021)
Article
Health Care Sciences & Services
Yali Qu, Haoyan Shang, Jing Li, Shenghua Teng
Summary: This study proposed a method to simplify sEMG devices by selecting informative channels through a combination of variable selection algorithms, which resulted in improved gesture recognition performance.
JOURNAL OF HEALTHCARE ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Keyu Liu, Tianrui Li, Xibei Yang, Hongmei Chen, Jie Wang, Zhixuan Deng
Summary: In this article, a novel semi-supervised feature selection scheme called SemiFREE is proposed, which redefines the feature relevance and redundancy by considering the fuzziness or uncertainty in data labeling. Experimental results demonstrate the superiority of SemiFREE in the presence of partially labeled data.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Utkarsh Agrawal, Vasudha Rohatgi, Rahul Katarya
Summary: The problem of feature selection involves selecting the most informative subset of features that have the most impact in classification. This paper proposes a novel variant of the Equilibrium Optimizer called Normalized Mutual Information-based equilibrium optimizer (NMIEO) for feature selection. The proposed method incorporates a local search strategy based on Normalized Mutual Information and utilizes Chaotic maps for population initialization. Experimental results demonstrate the superior performance of NMIEO compared to other competitive methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Chemistry, Multidisciplinary
Jerome Mendes, Ricardo Maia, Rui Araujo, Francisco A. A. Souza
APPLIED SCIENCES-BASEL
(2020)
Article
Chemistry, Multidisciplinary
Morad Danishvar, Sebelan Danishvar, Francisco Souza, Pedro Sousa, Alireza Mousavi
Summary: By utilizing real-time monitoring control technology and deep neural network models, the efficiency and stability of the cement processing grinding process have been improved, resulting in an increase in resource utilization efficiency.
APPLIED SCIENCES-BASEL
(2021)
Article
Agriculture, Multidisciplinary
T. Barros, P. Conde, G. Goncalves, C. Premebida, M. Monteiro, C. S. S. Ferreira, U. J. Nunes
Summary: This study presents a semantic segmentation approach for vine detection in real-world vineyards using state-of-the-art deep segmentation networks and conventional unsupervised methods. The results indicate that deep learning networks outperform classical methods in vine segmentation, and high-definition RGB images produce equivalent or higher performance than lower resolution multispectral band combinations.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Engineering, Electrical & Electronic
Jose Naranjo, Felipe Jimenez, Rodrigo Castineira, Mauro Gil, Cristiano Premebida, Pedro Serra, Alberto Valejo, Fawzi Nashashibi, Conceicao Magalhaes
IEEE INTELLIGENT TRANSPORTATION SYSTEMS MAGAZINE
(2022)
Article
Automation & Control Systems
Jorge S. S. Junior, Jerome Mendes, Francisco Souza, Cristiano Premebida
Summary: Deep learning has gained attention in regression applications, but often lacks interpretability. This paper investigates the state-of-the-art of deep fuzzy systems (DFS), which combine deep learning and fuzzy logic systems (FLS) for regression with good accuracy and interpretability. It emphasizes the importance of considering interpretability in the development of intelligent models.
INTERNATIONAL JOURNAL OF FUZZY SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Weihao Lu, Dezong Zhao, Cristiano Premebida, Li Zhang, Wenjing Zhao, Daxin Tian
Summary: This paper proposes a Point Augmentation (PA)-RCNN method for small object detection, which generates efficient complementary features without trainable parameters. Experimental results demonstrate its superiority in detecting distant and small objects compared to existing state-of-the-art methods.
IEEE TRANSACTIONS ON INTELLIGENT VEHICLES
(2023)
Article
Automation & Control Systems
Francisco Souza, Tim Offermans, Ruud Barendse, Geert Postma, Jeroen Jansen
Summary: This article presents a novel data-driven model, called contextual mixture of experts (cMoE), that incorporates process knowledge to enhance the synergy between humans and machines in the process industry. The cMoE utilizes process knowledge to shape historical data into possibility distributions, representing operators' context related to the process. The model is evaluated in two real case studies for quality prediction and demonstrates improved predictive performance and interpretability by providing insights into the variables affecting different process regimes.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Engineering, Civil
Gledson Melotti, Weihao Lu, Pedro Conde, Dezong Zhao, Alireza Asvadi, Nuno Goncalves, Cristiano Premebida
Summary: This paper proposes an approach to alleviate the problem of overconfident predictions in deep object detection networks by introducing a novel probabilistic layer. The approach reduces overconfidence in false positives without degrading the performance on true positives.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Information Systems
Gledson Melotti, Cristiano Premebida, Jordan J. Bird, Diego R. Faria, Nuno Goncalves
Summary: This study proposes a probabilistic approach based on pre-trained networks to improve object recognition. By utilizing distributions calculated from the Logit layer scores, new decision layers are constructed for Maximum Likelihood (ML) and Maximum a-Posteriori (MAP) inference. The ML and MAP layers show better performance and interpretable probabilistic predictions compared to Softmax and Sigmoid layers, as demonstrated on RGB images and LiDAR data.
Proceedings Paper
Automation & Control Systems
Weihao Lu, Dezong Zhao, Cristiano Premebida, Wen-Hua Chen, Daxin Tian
Summary: This paper proposes a deep neural network for 3D point cloud processing, emphasizing effective feature aggregation methods using fixed-radius grouping and spherical kernel convolution. The algorithm achieves state-of-the-art results and competitive accuracy in both classification and part segmentation tasks, while maintaining an efficient architecture.
2021 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2021)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Jack N. C. Hayton, Tiago Barros, Cristiano Premebida, Matthew J. Coombes, Urbano J. Nunes
2020 IEEE INTERNATIONAL CONFERENCE ON AUTONOMOUS ROBOT SYSTEMS AND COMPETITIONS (ICARSC 2020)
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Gledson Melotti, Cristiano Premebida, Nuno Goncalves
2020 IEEE INTERNATIONAL CONFERENCE ON AUTONOMOUS ROBOT SYSTEMS AND COMPETITIONS (ICARSC 2020)
(2020)
Article
Computer Science, Artificial Intelligence
Guang-Hai Liu, Zuo-Yong Li, Jing-Yu Yang, David Zhang
Summary: This article introduces a novel image retrieval method that improves retrieval performance by using sublimated deep features. The method incorporates orientation-selective features and color perceptual features, effectively mimicking these mechanisms to provide a more discriminating representation.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Fengguang Peng, Zihan Ding, Ziming Chen, Gang Wang, Tianrui Hui, Si Liu, Hang Shi
Summary: RGB-Thermal (RGB-T) semantic segmentation is an emerging task that aims to improve the robustness of segmentation methods under extreme imaging conditions by using thermal infrared modality. The challenges of foreground-background distinguishment and complementary information mining are addressed by proposing a cross modulation process with two collaborative components. Experimental results show that the proposed method achieves state-of-the-art performances on current RGB-T segmentation benchmarks.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Baihong Han, Xiaoyan Jiang, Zhijun Fang, Hamido Fujita, Yongbin Gao
Summary: This paper proposes a novel automatic prompt generation method called F-SCP, which focuses on generating accurate prompts for low-accuracy classes and similar classes. Experimental results show that our approach outperforms state-of-the-art methods on six multi-domain datasets.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Huikai Liu, Ao Zhang, Wenqian Zhu, Bin Fu, Bingjian Ding, Shengwu Xiong
Summary: Adverse weather conditions present challenges for computer vision tasks, and image de-weathering is an important component of image restoration. This paper proposes a multi-patch skip-forward structure and a Residual Deformable Convolutional module to improve feature extraction and pixel-wise reconstruction.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Oliver M. Crook, Mihai Cucuringu, Tim Hurst, Carola-Bibiane Schonlieb, Matthew Thorpe, Konstantinos C. Zygalakis
Summary: The transportation LP distance (TLP) is a generalization of the Wasserstein WP distance that can be applied directly to color or multi-channelled images, as well as multivariate time-series. TLP interprets signals as functions, while WP interprets signals as measures. Although both distances are powerful tools in modeling data with spatial or temporal perturbations, their computational cost can be prohibitively high for moderate pattern recognition tasks. The linear Wasserstein distance offers a method for projecting signals into a Euclidean space, and in this study, we propose linear versions of the TLP distance (LTLP) that show significant improvement over the linear WP distance in signal processing tasks while being several orders of magnitude faster to compute than the TLP distance.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Haitao Tian, Shiru Qu, Pierre Payeur
Summary: This paper proposes a method of target-dependent classifier, which optimizes the joint hypothesis of domain adaptation into a target-dependent hypothesis that better fits with the target domain clusters through an unsupervised fine-tuning strategy and the concept of meta-learning. Experimental results demonstrate that this method outperforms existing techniques in synthetic-to-real adaptation and cross-city adaptation benchmarks.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Qingsen Yan, Axi Niu, Chaoqun Wang, Wei Dong, Marcin Wozniak, Yanning Zhang
Summary: Deep learning-based methods have achieved remarkable results in the field of super-resolution. However, the limitation of paired training image sets has led researchers to explore self-supervised learning. However, the assumption of inaccurate downscaling kernel functions often leads to degraded results. To address this issue, this paper introduces KGSR, a kernel-guided network that trains both upscaling and downscaling networks to generate high-quality high-resolution images even without knowing the actual downscaling process.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Yifan Chen, Xuelong Li
Summary: Gait recognition is a popular technology for identification due to its ability to capture gait features over long distances without cooperation. However, current methods face challenges as they use a single network to extract both temporal and spatial features. To solve this problem, we propose a two-branch network that focuses on spatial and temporal feature extraction separately. By combining these features, we can effectively learn the spatio-temporal information of gait sequences.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Wei Shi, Wentao Zhang, Wei-shi Zheng, Ruixuan Wang
Summary: This article proposes a simple yet effective visualization framework called PAMI, which does not require detailed model structure and parameters to obtain visualization results. It can be applied to various prediction tasks with different model backbones and input formats.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Xiaobo Hu, Jianbo Su, Jun Zhang
Summary: This paper reviews the latest technologies in pattern recognition, highlighting their instabilities and failures in practical applications. From a control perspective, the significance of disturbance rejection in pattern recognition is discussed, and the existing problems are summarized. Finally, potential solutions related to the application of compensation on features are discussed to emphasize future research directions.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Andres Felipe Posada-Moreno, Nikita Surya, Sebastian Trimpe
Summary: Convolutional neural networks are widely used in critical systems, and explainable artificial intelligence has proposed methods for generating high-level explanations. However, these methods lack the ability to determine the location of concepts. To address this, we propose a novel method for automatic concept extraction and localization based on pixel-wise aggregations, and validate it using synthetic datasets.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Peng Bao, Jianian Li, Rong Yan, Zhongyi Liu
Summary: In this paper, a novel Dynamic Graph Contrastive Learning framework, DyGCL, is proposed to capture the temporal consistency in dynamic graphs and achieve good performance in node representation learning.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Kristian Schultz, Saptarshi Bej, Waldemar Hahn, Markus Wolfien, Prashant Srivastava, Olaf Wolkenhauer
Summary: Research indicates that deep generative models perform poorly compared to linear interpolation-based methods for synthetic data generation on small, imbalanced tabular datasets. To address this, a new approach called ConvGeN, combining convex space learning with deep generative models, has been proposed. ConvGeN improves imbalanced classification on small datasets while remaining competitive with existing linear interpolation methods.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Khondaker Tasrif Noor, Antonio Robles-Kelly
Summary: In this paper, the authors propose H-CapsNet, a capsule network designed for hierarchical image classification. The network effectively captures hierarchical relationships using dedicated capsules for each class hierarchy. A modified hinge loss is utilized to enforce consistency among the involved hierarchies. Additionally, a strategy for dynamically adjusting training parameters is presented to achieve better balance between the class hierarchies. Experimental results demonstrate that H-CapsNet outperforms competing hierarchical classification networks.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Lei Liu, Guorun Li, Yuefeng Du, Xiaoyu Li, Xiuheng Wu, Zhi Qiao, Tianyi Wang
Summary: This study proposes a new agricultural image segmentation model called CS-Net, which uses Simple-Attention Block and Simpleformer to improve accuracy and inference speed, and addresses the issue of performance collapse of Transformers in agricultural image processing.
PATTERN RECOGNITION
(2024)