Article
Computer Science, Information Systems
Huan Zhang, Liangxiao Jiang, Chaoqun Li
Summary: In this study, a novel model called A(2)WNB is proposed to address the limitation of the attribute conditional independence assumption in naive Bayes algorithm. By discovering and utilizing latent attributes beyond the original attribute space, as well as optimizing attribute weights to reduce attribute redundancy, the A(2)WNB model demonstrates superior performance in classification tasks.
SCIENCE CHINA-INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Huan Zhang, Liangxiao Jiang, Liangjun Yu
Summary: Naive Bayes algorithm remains one of the top 10 data mining algorithms, but its conditional independence assumption is rarely true in real-world applications. This study introduces a new improved model called AIWNB, which combines attribute weighting and instance weighting into one uniform framework, to address this issue effectively.
PATTERN RECOGNITION
(2021)
Article
Computer Science, Artificial Intelligence
Huan Zhang, Liangxiao Jiang, Chaoqun Li
Summary: The proposed collaboratively weighted Naive Bayes (CWNB) approach outperforms the standard NB and all other existing state-of-the-art competitors by simultaneously learning instance and attribute weights in a collaborative manner.
KNOWLEDGE AND INFORMATION SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Taeheung Kim, Jong-Seok Lee
Summary: Imbalanced data classification is a common problem in real-world applications, and traditional classification algorithms are hindered by their focus on overall accuracy. To address this challenge, we proposed a novel attribute weighting method called RankOptAUC NB (RNB) for naive Bayes classifier. RNB formulates learning a weighted NB classifier as a nonlinear optimization problem to maximize the area under the ROC (AUC) curve. The proposed method improves the performance of weighted NB classifier by selecting important variables using a regularization term added to the objective function. Experimental results using 30 real-world datasets demonstrate the effectiveness of the proposed approach in determining optimal attribute weights for imbalanced data classification.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Huan Zhang, Liangxiao Jiang
Summary: Naive Bayes (NB) is a simple, efficient, and effective data mining algorithm. However, its performance is limited by the unrealistic attribute conditional independence assumption and unreliable conditional probability estimation. This study proposes a novel model called fine tuned attribute weighted NB (FTAWNB), which combines fine tuning with attribute weighting to enhance NB's performance by improving both the attribute conditional independence assumption and conditional probability estimation.
Article
Computer Science, Artificial Intelligence
Huan Zhang, Liangxiao Jiang, Wenjun Zhang, Chaoqun Li
Summary: In this study, a new model called multi-view attribute weighted naive Bayes (MAWNB) is proposed to portray data characteristics more comprehensively. By constructing two label views from raw attributes and optimizing attribute weights, MAWNB can predict class labels for test instances with high accuracy. Extensive experiments demonstrate the superiority of MAWNB compared to NB and other state-of-the-art competitors.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Chemistry, Multidisciplinary
Fahad S. Alenazi, Khalil El Hindi, Basil AsSadhan
Summary: This study proposes a fine-grained boosting method for improving the performance of the Naive Bayes classifier by identifying hidden attribute values that cause underfitting or overfitting. The method captures the correlation between attribute values and their impact on the model's performance. Empirical results show that the proposed method outperforms existing approaches.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Taeheung Kim, Jong-Seok Lee
Summary: This study proposes attribute weighting methods for naive Bayes classification to improve the classification performance by minimizing different loss functions. The proposed loss functions belong to exponential functions, making the optimization problems easier to solve and allowing for easy modification to enhance the classifier's robustness to noisy instances. Experimental results confirm the effectiveness of the proposed scheme.
Article
Computer Science, Artificial Intelligence
Shufen Ruan, Baozhou Chen, Kunfang Song, Hongwei Li
Summary: This paper presents an innovative attribute weighting method for naive Bayes text classifiers, utilizing an improved distance correlation coefficient to accurately measure the importance of attributes to categories, achieving optimization. Experimental results indicate an effective balance between classification accuracy and execution time with this method.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Xiaoliang Zhou, Donghua Wu, Zitong You, Dongyang Wu, Ning Ye, Li Zhang
Summary: This paper introduces an adaptive two-index fusion attribute-weighted NB (ATFNB) method to overcome the challenges caused by the attribute independence assumption in the Naive Bayes algorithm and improve accuracy.
Article
Computer Science, Artificial Intelligence
Shihe Wang, Jianfeng Ren, Ruibin Bai
Summary: Recently, improved naive Bayes methods, including regularized naive Bayes (RNB), have been developed to enhance discrimination capabilities. However, these methods often result in significant information loss due to inadequate data discretization. To address this issue, we propose a semi-supervised adaptive discriminative discretization framework that utilizes both labeled and unlabeled data to better estimate the data distribution. Our proposed method, called RNB+, shows superior performance compared to state-of-the-art NB classifiers by significantly reducing information loss and improving discrimination power.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Lee-Kien Foo, Sook-Ling Chua, Neveen Ibrahim
Summary: The naive Bayes classifier is a simple yet effective method for data mining classification. However, the assumption of attribute independence may not hold in real-world applications. To address this, researchers proposed a method to incorporate attribute weights into naive Bayes, which resulted in improved classification performance in terms of accuracy and F1 score.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Article
Computer Science, Artificial Intelligence
Huan Zhang, Liangxiao Jiang, Geoffrey I. Webb
Summary: Naive Bayes is a classical machine learning algorithm that often uses discretization to transform quantitative attributes into qualitative attributes. Non-Disjoint Discretization (NDD) is a novel method that forms overlapping intervals and always locates a value toward the middle of an interval. However, existing approaches to NDD fail to consider the effect of multiple occurrences of a single value. In this study, a new method called Rigorous Non-Disjoint Discretization (RNDD) is proposed to handle multiple occurrences of a single value in a systematic manner, and it outperforms NDD and other existing competitors.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Information Systems
Fang Gong, Xin Wang, Liangxiao Jiang, Seyyed Mohammadreza Rahimi, Dianhong Wang
Summary: A novel fine-grained attribute weighting scheme is proposed in this study to optimize the inverted specific-class distance measure, which significantly outperforms the original ISCDM and other competing methods in terms of negative conditional log likelihood and root relative squared error on machine learning datasets.
INFORMATION SCIENCES
(2021)
Article
Mathematics
Liangjun Yu, Shengfeng Gan, Yu Chen, Dechun Luo
Summary: This paper introduces Naive Bayes and its improved model IWHNB, which combines the improved HNB model with instance weights to achieve significant improvements in classification performance.
Article
Computer Science, Information Systems
Yu Dong, Liangxiao Jiang, Chaoqun Li
Summary: Crowdsourcing enables faster and cheaper label acquisition for supervised learning, but noise in the labels can degrade model performance. To address this, a novel co-training-based noise correction algorithm (CTNC) is proposed, which outperforms all current state-of-the-art label noise correction algorithms.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Ziqi Chen, Liangxiao Jiang, Chaoqun Li
Summary: In crowdsourcing scenarios, obtaining multiple noisy labels for each instance and integrating them is a common practice. However, there is still label noise in the integrated labels. To mitigate the impact of label noise, researchers have proposed various noise correction methods. This paper introduces a simple yet effective noise correction method called LDNC for multiclass classification tasks, which outperforms existing state-of-the-art methods.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Ben Ma, Chaoqun Li, Liangxiao Jiang
Summary: This paper proposes a novel ground truth inference algorithm based on instance similarity to improve the performance of ground truth inference by utilizing the similarity between instances.
APPLIED INTELLIGENCE
(2022)
Article
Automation & Control Systems
Lijuan Ren, Liangxiao Jiang, Chaoqun Li
Summary: In crowdsourcing scenarios, a label confidence-based noise correction (LCNC) method is proposed to filter noisy instances and correct label errors. LCNC first calculates the confidence of each instance's noisy labels to separate clean and noise sets. Then, it utilizes random trees to recalculate the label confidence and correct the noise instances using a consensus voting strategy.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Yufei Hu, Liangxiao Jiang, Chaoqun Li
Summary: This paper proposes a noise correction algorithm called instance difficulty-based noise correction (IDNC), which takes into account the effect of instance difficulty on noise correction. Experimental results demonstrate the effectiveness and efficiency of IDNC on both simulated and real-world crowdsourced datasets.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Wenjun Yang, Chaoqun Li, Liangxiao Jiang
Summary: The crowdsourcing system provides an easy way to obtain labeled training data, but the quality of labels provided by non-expert labelers tends to be low. This paper proposes a new ground truth inference algorithm, called RSVMI, which addresses the learning-from-crowds problem from the perspective of robust classifiers. The experimental results demonstrate the effectiveness of RSVMI on benchmark and real-world datasets, especially when the number of labelers is very small.
SCIENCE CHINA-INFORMATION SCIENCES
(2023)
Article
Computer Science, Information Systems
Yao Zhang, Liangxiao Jiang, Chaoqun Li
Summary: Crowdsourcing is an effective and low-cost method for collecting labels, however, the quality of these labels is often low due to the insufficient professional knowledge of crowd workers. To address this issue, this paper proposes a novel three-stage label integration method called Attribute Augmentation-based Label Integration (AALI).
FRONTIERS OF COMPUTER SCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Xinyang Li, Chaoqun Li, Liangxiao Jiang
Summary: This paper proposes a novel multi-view-based noise correction algorithm (MVNC) that utilizes the idea of multi-view learning to correct noise instances in crowdsourced data. Experimental results demonstrate that the new view significantly enhances the effect of noise correction.
INFORMATION FUSION
(2023)
Article
Computer Science, Information Systems
Huiru Li, Liangxiao Jiang, Siqing Xue
Summary: In this article, a neighborhood weighted voting-based noise correction algorithm (NWVNC) is proposed, which utilizes the multiple noisy label sets of each instance's neighbors to estimate the probability that it belongs to its integrated label. The algorithm then filters noise instances and corrects them by the consensus voting of three trained classifiers. Experimental results demonstrate that NWVNC outperforms other state-of-the-art noise correction algorithms.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2023)
Article
Computer Science, Artificial Intelligence
Kang Zhu, Siqing Xue, Liangxiao Jiang
Summary: In the crowdsourcing scenario, repeated labeling is used to obtain noisy label sets for each instance, which are then integrated using a ground truth inference method. However, integrated labels still contain noise. Many noise correction methods have been proposed, but they filter out too many instances and can only use a few clean instances for learning. In this paper, we propose a two-stage noise correction method called DCTNC, which learns from both clean and noise instances and achieves state-of-the-art results.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2023)
Article
Computer Science, Information Systems
Wenbin Li, Chaoqun Li, Liangxiao Jiang
Summary: Crowdsourcing systems are useful for obtaining data labels. The challenge is to infer integrated labels from multiple crowd labels, especially when there are only a few labels. This paper proposes robust logistic regression inference (RLRI) algorithms to handle noise in multiple crowd labels and achieve good performance with a small number of labels.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Xue Wu, Liangxiao Jiang, Wenjun Zhang, Chaoqun Li
Summary: In crowdsourcing learning, label noise exists in the integrated labels obtained by label integration algorithms. To decrease the impact of label noise, many scholars focus on noise correction algorithms. The proposed TDNC algorithm, inspired by three-way decision theory, outperforms existing noise correction algorithms in terms of noise ratio.
INTERNATIONAL JOURNAL OF APPROXIMATE REASONING
(2023)
Article
Computer Science, Artificial Intelligence
Huan Zhang, Liangxiao Jiang, Geoffrey I. Webb
Summary: Naive Bayes is a classical machine learning algorithm that often uses discretization to transform quantitative attributes into qualitative attributes. Non-Disjoint Discretization (NDD) is a novel method that forms overlapping intervals and always locates a value toward the middle of an interval. However, existing approaches to NDD fail to consider the effect of multiple occurrences of a single value. In this study, a new method called Rigorous Non-Disjoint Discretization (RNDD) is proposed to handle multiple occurrences of a single value in a systematic manner, and it outperforms NDD and other existing competitors.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Information Systems
Qiang Ji, Liangxiao Jiang, Wenjun Zhang
Summary: In this article, a dual-view noise correction (DVNC) algorithm is proposed, which enhances the effect of noise correction by fully utilizing the joint information of the original attribute view and multiple noisy label view.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Computer Science, Artificial Intelligence
Huan Zhang, Liangxiao Jiang, Wenjun Zhang, Chaoqun Li
Summary: In this study, a new model called multi-view attribute weighted naive Bayes (MAWNB) is proposed to portray data characteristics more comprehensively. By constructing two label views from raw attributes and optimizing attribute weights, MAWNB can predict class labels for test instances with high accuracy. Extensive experiments demonstrate the superiority of MAWNB compared to NB and other state-of-the-art competitors.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
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)