TSK fuzzy system fusion at sensitivity-ensemble-level for imbalanced data classification

Zhang, Yuanpeng; Wang, Guanjin; Huang, Xiuyu; Ding, Weiping

doi:10.1016/j.inffus.2022.12.014

Cited by 28 publications

(7 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the field of artificial intelligence, this aspect is particularly important in decision-making problems under uncertainty and complexity, such as those in financial analysis. Fuzzy sets are used in various applications, such as control systems [54], data classification [55], and expert modeling [56]. By their ability to capture nuances and ambiguities in real data, fuzzy sets have become an essential tool in the field of AI, helping to improve the accuracy and robustness of intelligent systems.…”

Section: Discussionmentioning

confidence: 99%

Mathematical Patterns in Fuzzy Logic and Artificial Intelligence for Financial Analysis: A Bibliometric Study

Nica,

Delcea,

Chiriță

2024

Mathematics

View full text Add to dashboard Cite

In this study, we explored the dynamic field of fuzzy logic and artificial intelligence (AI) in financial analysis from 1990 to 2023. Utilizing the bibliometrix package in RStudio and data from the Web of Science, we focused on identifying mathematical models and the evolving role of fuzzy information granulation in this domain. The research addresses the urgent need to understand the development and impact of fuzzy logic and AI within the broader scope of evolving technological and analytical methodologies, particularly concentrating on their application in financial and banking contexts. The bibliometric analysis involved an extensive review of the literature published during this period. We examined key metrics such as the annual growth rate, international collaboration, and average citations per document, which highlighted the field’s expansion and collaborative nature. The results revealed a significant annual growth rate of 19.54%, international collaboration of 21.16%, and an average citation per document of 25.52. Major journals such as IEEE Transactions on Fuzzy Systems, Fuzzy Sets and Systems, the Journal of Intelligent & Fuzzy Systems, and Information Sciences emerged as significant contributors, aligning with Bradford’s Law’s Zone 1. Notably, post-2020, IEEE Transactions on Fuzzy Systems showed a substantial increase in publications. A significant finding was the high citation rate of seminal research on fuzzy information granulation, emphasizing its mathematical importance and practical relevance in financial analysis. Keywords like “design”, “model”, “algorithm”, “optimization”, “stabilization”, and terms such as “fuzzy logic controller”, “adaptive fuzzy controller”, and “fuzzy logic approach” were prevalent. The Countries’ Collaboration World Map indicated a strong pattern of global interconnections, suggesting a robust framework of international collaboration. Our study highlights the escalating influence of fuzzy logic and AI in financial analysis, marked by a growth in research outputs and global collaborations. It underscores the crucial role of fuzzy information granulation as a mathematical model and sets the stage for further investigation into how fuzzy logic and AI-driven models are transforming financial and banking analysis practices worldwide.

show abstract

Section: Discussionmentioning

confidence: 99%

Mathematical Patterns in Fuzzy Logic and Artificial Intelligence for Financial Analysis: A Bibliometric Study

Nica,

Delcea,

Chiriță

2024

Mathematics

View full text Add to dashboard Cite

show abstract

“…In classification tasks, CIP emerges when the number of samples in one class (majority class) is significantly larger than that in other classes (minority classes) [87]. CIP can substantially affect the performance of ML models because it leads to bias during learning algorithms, emphasizing the classification of samples of the majority class and the misclassification of samples of minority classes [21].…”

Section: Data Augmentation Techniques For Imbalanced Datamentioning

confidence: 99%

Ensemble feature selection and tabular data augmentation with generative adversarial networks to enhance cutaneous melanoma identification and interpretability

Gómez-Martínez,

Chushig-Muzo,

Veierød

et al. 2024

Preprint

View full text Add to dashboard Cite

Background: Cutaneous melanoma, the most serious skin cancer, constitutes a considerable health burden in fair-skinned populations. Its increasing incidence highlights the need to develop automated computer-aided approaches that support dermatologists in melanoma detection. Recent advances in artificial intelligence (AI), jointly with the availability of public dermoscopy image datasets, have led to create data-driven models with high predictive performance, thus contributing to dermatology research. Despite these advantages, most image datasets present the class imbalance problem, where a few classes have numerous samples, whereas others are under-represented, impacting the performance of AI-based models. Methods: In this paper, we propose to combine ensemble feature selection (FS) methods and data augmentation with the conditional tabular generative adversarial networks (CTGAN) to enhance melanoma identification in imbalanced datasets. We employed dermoscopy images from two public datasets, PH2 and Derm7pt, which contain melanoma and not-melanoma lesions. To capture intrinsic information from skin lesions, we conduct two feature extraction (FE) approaches, including handcrafted and embedding features. For the former, color, geometric and first-, second-, and higher-order texture features were extracted, while for the latter, embeddings were obtained using ResNet-based models. To alleviate the high-dimensionality in the FE, ensemble FS with filter methods were used and evaluated. For data augmentation, we conducted a progressive analysis of the imbalance ratio (IR), related to the amount of synthetic samples created, and evaluated the impact on the predictive results. To gain interpretability on predictive models, we used SHAP, bootstrap resampling statistical tests and UMAP visualizations. Results: Experimental results showed that the combination of CTGAN and linear models achieved the best predictive results, achieving AUCROC values of 86% (with LASSO and IR=0.8) and 71% (with support vector machine and IR=0.9) for the PH2 and Derm7pt, respectively. We also identified that ‘melanoma’ lesions were mainly characterized by features related to color, while ‘not melanoma’ lesions were characterized by texture features. Conclusions: Our results demonstrated the effectiveness of synthetic data to build generalizable models for melanoma prediction. Our work contributes to skin lesion research, helping in melanoma identification and the interpretation of main characteristics associated with melanoma.

show abstract

“…The main benchmarking platforms are CIFAR-10, CIFAR-100 ( Krizhevsky, 2009 ), TinyImageNet ( Le & Yang, 2015 ), ImageNet-1K ( Russakovsky et al, 2015 ), etc . The evaluation indicators are depth, width, params, FLOPs, and accuracy ( Zhang et al, 2023 ). Some previous state-of-the-art network architectures are selected for comparison, such as ResNets ( He et al, 2016b ), RepVGG ( Ding et al, 2021 ), DenseNet ( Huang et al, 2017 ), ResNexts ( Xie et al, 2017 ), WideResNet ( Zagoruyko & Komodakis, 2016 ), SENet ( Hu et al, 2020 ), ParNet ( Goyal et al, 2021 ), MobileV2 ( Sandler et al, 2018 ), EfficientNet ( Tan & Le, 2019 ), EfficientNetV2 ( Tan & Le, 2021 ), etc .…”

Section: Benchmarkingmentioning

confidence: 99%

TbsNet: the importance of thin-branch structures in CNNs

Sheng

Shi

et al. 2023

PeerJ Computer Science

View full text Add to dashboard Cite

The performance of a convolutional neural network (CNN) model is influenced by several factors, such as depth, width, network structure, size of the receptive field, and feature map scaling. The optimization of the best combination of these factors poses as the main difficulty in designing a viable architecture. This article presents an analysis of key factors influencing network performance, offers several strategies for constructing an efficient convolutional network, and introduces a novel architecture named TbsNet (thin-branch structure network). In order to minimize computation costs and feature redundancy, lightweight operators such as asymmetric convolution, pointwise convolution, depthwise convolution, and group convolution are implemented to further reduce the network’s weight. Unlike previous studies, the TbsNet architecture design rejects the reparameterization method and adopts a plain, simplified structure which eliminates extraneous branches. We conduct extensive experiments, including network depth, width, etc. TbsNet performs well on benchmark platforms, Top 1 Accuracy on CIFAR-10 is 97.02%, on CIFAR-100 is 83.56%, and on ImageNet-1K is 86.17%. Tbs-UNet’s DSC on the Synapse dataset is 78.39%, higher than TransUNet’s 0.91%. TbsNet can be competent for some downstream tasks in computer vision, such as medical image segmentation, and thus is competitive with prior state-of-the-art deep networks such as ResNet, ResNeXt, RepVgg, ParNet, ConvNeXt, and MobileNet.

show abstract

TSK fuzzy system fusion at sensitivity-ensemble-level for imbalanced data classification

Cited by 28 publications

References 35 publications

Mathematical Patterns in Fuzzy Logic and Artificial Intelligence for Financial Analysis: A Bibliometric Study

Mathematical Patterns in Fuzzy Logic and Artificial Intelligence for Financial Analysis: A Bibliometric Study

Ensemble feature selection and tabular data augmentation with generative adversarial networks to enhance cutaneous melanoma identification and interpretability

TbsNet: the importance of thin-branch structures in CNNs

Contact Info

Product

Resources

About