检索结果-内蒙古大学图书馆

BERT-PhishFinder: A Robust Model for Accurate Phishing URL Detection with Optimized DistilBERT

IEEE Transactions on Dependable and Secure Computing 2025年

作者： Aljofey, Ali Bello, Saifullahi Aminu Lu, Jian Xu, Chen Shenzhen University Shenzhen Key Laboratory of Advanced Machine Learning and Applications College of Mathematics and Statistics Shenzhen518060 China Shenzhen518055 China

Phishing URL detection has become a critical challenge in cybersecurity, with existing methods often struggling to maintain high accuracy while generalizing across diverse datasets. In this paper, we introduce BERT-PhishFinder, a novel and efficient transformer-based model designed to tackle this problem. While most traditional approaches rely heavily on lexical features or complex convolutional architectures, BERT-PhishFinder leverages the power of DistilBERT, a lightweight yet highly effective transformer, to capture rich contextual representations of URL sequences. To enhance the model's robustness and reduce overfitting, we strategically incorporate SpatialDropout1D in the embedding layers, along with global average pooling and global max pooling techniques to extract both comprehensive and key discriminative features. The pooled representations are thoughtfully concatenated to form a comprehensive feature representation. Through this carefully crafted design, our model adopts ensemble learning, as it undergoes multiple parallel dense layers, each with distinct parameters and dropout regularization. This facilitates learning diverse patterns and features from the input URL sequence, culminating in exceptional phishing URL detection performance. Extensive evaluations against conventional deep learning algorithms, transformer models (XLNet, RoBERTa, ALBERT), and other existing methods on five benchmark datasets show that BERT-PhishFinder not only achieves the state of-the-art real phishing URL detection but also accomplishes this with reduced label dependency. © 2004-2012 IEEE.

关键词： Phishing

来源：评论

学校读者我要写书评

暂无评论

CRLNet: Cascaded Resolution learning Network for Natural Scenes Segmentation

引用

IEEE Intelligent Systems 2025年

作者： Li, Wei Tian, Shishun Hua, Guoguang Liao, Muxin Zhang, Yuhang Zou, Wenbin Shenzhen University Guangdong Key Laboratory of Intelligent Information Processing Shenzhen Key Laboratory of Advanced Machine Learning and Applications College of Electronics and Information Engineering Shenzhen518060 China Jiangxi Agricultural University School of Computer Science and Engineering Nanchang330045 China

The natural environment presents a multitude of scenes with diverse content, posing challenges for satisfactory segmentation results using existing segmentation networks. In response, we propose a Cascaded Resolution learning Network (CRLNet) to enhance segmentation performance through global textual embedding and multi-resolution feature learning. The CRLNet constructs a multi-path segmentation system that integrates multi-resolution feature data from different paths, thereby progressively enhancing local feature learning. Two key modules, the Partition-Fusion Channel Attention Module (PFCAM) and the Features learning Module (FLM), are pivotal components of CRLNet. The PFCAM serves as a computationally efficient channel attention module to mitigate segmentation confusion stemming from similar objects. Meanwhile, the FLM is tailored to learn resolution feature maps from different paths, facilitating the refinement of object representation and enhancing segmentation performance. Extensive experiments conducted on real natural scene datasets demonstrate the superiority of the proposed CRLNet over existing efficient segmentation methods in terms of accuracy. © 2025 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

DECOMPOSITION NUMBERS OF CYCLOTOMIC BRAUER ALGEBRAS OVER THE COMPLEX FIELD, I

arXiv

引用

arXiv 2025年

作者： Gao, Mengmeng Rui, Hebing Xiao, Wei School of Mathematical Science Tongji University Shanghai200092 China College of Mathematics and Statistics Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen University Guangdong Shenzhen518060 China

Following Nazarov’s suggestion [23], we refer to the cyclotomic Nazarov-Wenzl algebra as the cyclotomic Brauer algebra. When the cyclotomic Brauer algebra is isomorphic to the endomorphism algebra of MIi,r– the tensor product of a simple scalar-type parabolic Verma module with the natural module in the parabolic BGG category O of types Bn, Cn and Dn, its decomposition numbers can theoretically be computed, based on general results from [1] and [29, Corollary 5.10]. This paper aims to establish explicit connections between the parabolic Verma modules that appear as subquotients of MIi,r and the right cell modules of the cyclotomic Brauer algebra under condition (1.12). It allows us to explicitly decompose MIi,r into a direct sum of indecomposable tilting modules by identifying their highest weights and multiplicities. Our result demonstrates that the decomposition numbers of such a cyclotomic Brauer algebra can be explicitly computed using the parabolic Kazhdan-Lusztig polynomials of types Bn, Cn, and Dn with suitable parabolic subgroups [32]. Finally, condition (1.12) is well-supported by a result of Wei Xiao presented in Section 6. Copyright © 2025, The Authors. All rights reserved.

关键词： Polynomials

来源：评论

学校读者我要写书评

暂无评论

A novel energy functional minimization model for speckle noise removal

引用

Optoelectronics Letters 2019年第5期15卷 386-390页

作者： CHEN Bo ZOU Jin-bin CHEN Wen-sheng ZHANG Meng-yun Shenzhen Key Laboratory of Advanced Machine Learning and Applications College of Mathematics and Statistics Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University

In this paper, a novel energy functional minimization model is proposed for ultrasound images denoising. A controllable regularized term and the variational method are employed in the process of speckle noise. This model not only improves the plasticity of the model, but also improves the effect and efficiency of noise removal. The new model has different diffusion performance in different regions. At the same time, the diffusion performance is related to the parameters introduced by the proposed model. Numerical simulation results show that different parameters have different denoising effects, and the proposed model for speckle noise removal is superior to traditional models.

关键词： functional minimization model speckle noise removal ultrasound images denoising

来源：评论

学校读者我要写书评

暂无评论

Power control in d2d underlay distributed antenna systems 2nd

Power control in d2d underlay distributed antenna systems

引用

2nd International Conference on Sensor Networks and Signal Processing, SNSP 2019

作者： Qian, Gongbin Zhang, Ce He, Chunlong Li, Xingquan Tian, Chu Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen University Shenzhen518060 China

ISBN: (纸本)9789811549168

A new scenario is considered that device-to-device (D2D) communication users underlay the spectrum resource of cellular user in distributed antenna systems (DAS) is discussed in this paper. We mainly focus on how to improve spectral efficiency (SE) and energy efficiency (EE) of the system. Under the maximum transmit power constraint per antenna unit, we propose two resource allocation algorithms to solve the optimal problems of maximum SE and EE. The first problem can be transformed into a difference of convex (DC) structure problem by function recombination, then the concave-convex procedure (CCCP) algorithm and the interior point method which are adopted to get the optimal solutions for the maximum SE. Subsequently, by using the Dinkelbach algorithm based on the parameter method, a power allocation algorithm for energy efficiency is developed to solve the maximum EE optimization problem. The optimal solutions are also obtained by the CCCP algorithm and the interior point method. Simulation results show that compared to co-located antenna systems (CAS) with D2D users, the SE and EE performances of the proposed system have a significant improvement. © Springer Nature Singapore Pte Ltd 2021.

关键词： Spectrum efficiency

来源：评论

学校读者我要写书评

暂无评论

ON LOCAL CONTROLLABILITY FOR COMPRESSIBLE NAVIER-STOKES EQUATIONS WITH DENSITY DEPENDENT VISCOSITIES

引用

Acta Mathematica Scientia 2023年第2期43卷 675-685页

作者：连祥凯陶强姚正安 College of Mathematics Sun Yat-sen UniversityGuangzhou 510275China College of Mathematics and Statistics Shenzhen UniversityShenzhen 518060China Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen UniversityShenzhen 518060China

In this paper,we study the controllability of compressible Navier-Stokes equations with density dependent *** when the shear viscosityμis a positive constant and the bulk viscosityλis a function of the density,it is proven that the system is exactly locally controllable to a constant target trajectory by using boundary control functions.

关键词： compressible Navier-Stokes equations controllability density dependent vis-cosities

来源：评论

学校读者我要写书评

暂无评论

Contextual-aware terrain segmentation network for navigable areas with triple aggregation

引用

Expert Systems with applications 2025年

作者： Wei Li Muxin Liao Wenbin Zou Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen University 518060 Guangdong shenzhen China College of Electronics and Information Engineering Shenzhen University 518060 Guangdong shenzhen China School of Electronics and Information Engineering Jiangxi Agricultural University 330045 Jiangxi Nanchang China Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University 518060 Guangdong shenzhen China

Robust terrain feature representation is beneficial for accurately identifying navigable areas in autonomous vehicle systems, especially in complex and unstructured wild terrain environments. However, existing methods struggle to achieve stable and precise feature representation due to their inability to capture sufficient terrain features. To address these issues, we propose the Contextual-Aware Terrain Segmentation Network (CTSNet) with triple aggregation to effectively capture terrain features for safe navigable area segmentation. CTSNet comprises three essential modules: the Global Channel Sparsification Module (GCSM), the Fusion-wise Calibration Module (FCM), and the Attention Channel Entropy Module (ACEM). First, GCSM selectively globalizes channels through Top-K sparsification, enabling the aggregation of rich global contextual cues that guide feature fusion. Second, FCM refines multi-scale feature aggregation by addressing inconsistent feature representations across scales. It explores the spatial-semantic relationships between features of different sizes and enhances feature interaction via channel-wise calibration, ensuring more coherent integration of both fine-grained and global information. Finally, ACEM aggregates meaningful channel features through normalization and information entropy, ensuring that only the most relevant features contribute to the final segmentation output. Through triple aggregation in global semantic, multi-scale feature, and channel entropy, CTSNet obtains satisfactory and contextual-aware terrain features. Extensive experiments on multiple wild terrain datasets demonstrate that CTSNet outperforms state-of-the-art methods in identifying navigable areas. The code will be available at https://***/lv881314/CTSNet .

关键词：

来源：评论

学校读者我要写书评

暂无评论

Non-Local and Fully Connected Tensor Network Decomposition for Remote Sensing Image Denoising

引用

高等学校计算数学学报（英文版） 2024年第2期17卷 379-403页

作者： Zhihui Tu Shunda Chen Jian Lu Lin Li Qingtang Jiang Shenzhen Key Laboratory of Advanced Machine Learning and Applications School of Mathematical SciencesShenzhen UniversityShenzhen 518060China Shenzhen Key Laboratory of Advanced Machine Learning and Applications School of Mathematical SciencesShenzhen UniversityShenzhen 518060China National Center for Applied Mathematics Shenzhen(NCAMS) Shenzhen 518055China Pazhou Lab Guangzhou 510320China School of Electronic Engineering Xidian UniversityXi'an 710071China Department of Mathematics and Statistics University of Missouri-St.LouisSt.LouisMO 63121USA

Remote sensing images(RSIs)encompass abundant spatial and spec-tral/temporal information,finding wide applications in various ***,during image acquisition and transmission,RSI often encounter noise interference,which adversely affects the accuracy of subsequent *** address this is-sue,this paper proposes a novel non-local fully connected tensor network(NLFCTN)decomposition algorithm for denoising RSI,aiming to fully exploit their global cor-relation and non-local self-similarity(NSS)***,as a recently de-veloped tensor decomposition technique,exhibits remarkable capability in captur-ing global correlations and minimizing information *** addition,we introduce an efficient algorithm based on proximal alternating minimization(PAM)to effi-ciently solve the model and prove the *** effectiveness of the pro-posed method is validated through denoising experiments on both simulated and real RSI data,employing objective evaluation metrics and subjective visual *** results of the experiment show that the proposed method outperforms other RSI denoising techniques in terms of denoising performance.

关键词： Remote sensing images denoising tensor decomposition non-local self-similarity

来源：评论

学校读者我要写书评

暂无评论

Automatic inspection of yarn locations by utilizing histogram segmentation and monotone hypothesis 1

引用

2nd Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2019

作者： Han, Yu Luo, Ling Shenzhen Key Laboratory of Advanced Machine Learning and Applications College of Mathematics and Statistics Shenzhen University Shenzhen518060 China

ISBN: (数字)9783030317232

ISBN: (纸本)9783030317225

We present an automatic method for the estimation of yarn locations in fabric images. The proposed method is based on histogram segmentation and a so-called monotone hypothesis which is a nonparametric statistical approach. In this method, accumulated partial derivatives histograms are statistically unimodal, which indicate the periodic structures of fabric images. Then, the monotone hypothesis is applied to divide the histograms into several segments. According to the maximum value and the minimum value of the histograms in each segment, the locations of yarn boundaries and yarn centers can be correspondingly estimated. The method reduces the influence of yarn random texture noise that comes from yarn hairiness, improving the accuracy of detection. Furthermore, compared with classical method based on image smoothing, the proposed method can avoid over-smoothing of the edges of yarns. © Springer Nature Switzerland AG 2019.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

A Weighted Non-convex Restoration Model for Hyperspectral Image 11th

A Weighted Non-convex Restoration Model for Hyperspectral Im...

引用

11th International Conference on Image and Graphics, ICIG 2021

作者： Yu, Yijun Li, Min College of Mathematics and Statistics Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen University Shenzhen518060 China

ISBN: (纸本)9783030873578

Low-rank and sparsity are two popular regularization objects in hyperspectral image restoration. To generalize this decomposition, a weighted non-convex low-rank method is proposed by adopting non-convex lp -norm (0 p -norm can better approach the l0 -norm when 0 p -norm to the bases of the gradient of data. To remain faithful to the dissimilarity between the correlation along spatial and spectral directions, weights are assigned when computing the difference in each direction. Although being non-convex, the proposed model can be solved via ADMM algorithm. Extensive simulated experiments show that this model outperforms various existing methods visually and quantitatively. © 2021, Springer Nature Switzerland AG.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：