检索结果-南通市图书馆

Privacy-preserving explainable AI: a survey

同方学位论文库评论

在线全文

同方学位论文库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2025年第1期 23-56页

作者： thanh Tam NGUYEN thanh Trung huynh Zhao REN thanh Toan NGUYEN Phi Le NGUYEN Hongzhi YIN Quoc Viet Hung NGUYEN School of Information and Communication Technology Griffith University School of Computer and Communication Sciences Ecole Polytechnique Federale de Lausanne Faculty of Mathematics and Computer Science University of Bremen Faculty of Information Technology HUTECH University Department of Computer Science Hanoi University of Science and Technology School of Electrical Engineering and Computer Science The University of Queensland

As the adoption of explainable AI(XAI) continues to expand, the urgency to address its privacy implications intensifies. Despite a growing corpus of research in AI privacy and explainability, there is little attention on privacy-preserving model explanations. This article presents the first thorough survey about privacy attacks on model explanations and their countermeasures. Our contribution to this field comprises a thorough analysis of research papers with a connected taxonomy that facilitates the categorization of privacy attacks and countermeasures based on the targeted explanations. This work also includes an initial investigation into the causes of privacy leaks. Finally, we discuss unresolved issues and prospective research directions uncovered in our analysis. This survey aims to be a valuable resource for the research community and offers clear insights for those new to this domain. To support ongoing research, we have established an online resource repository, which will be continuously updated with new and relevant findings.

关键词：

Electromechanical admittance-based automatic damage assessment in plate structures via one-dimensional CNN-based deep learning models

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Structural and Civil Engineering 2024年第11期18卷 1730-1751页

作者： thanh-Canh huynh Nhat-Duc HOANG Quang-Quang PHAM Gia Toai TRUONG thanh-Truong NGUYEN Institute of Research and Development Duy Tan UniversityDa Nang 550000Vietnam Faculty of Civil Engineering Duy Tan UniversityDa Nang 550000Vietnam Bridge and Road Department Danang Architecture UniversityDa Nang 550000Vietnam Faculty of Civil Engineering and Technology Dong A UniversityDa Nang 550000Vietnam Industrial Maintenance Training Center Ho Chi Minh City University of Technology(HCMUT)Ho Chi Minh City 700000Vietnam

The conventional admittance approach utilizing statistical evaluation metrics offers limited information about the damage location,especially when damage introduces nonlinearities in admittance *** study proposes a novel automated damage localization method for plate-like structures based on deep learning of raw admittance signals.A one-dimensional(1D)convolutional neural network(CNN)-based model is designed to automate processing of raw admittance response and prediction of damage probabilities across multiple locations in a monitored *** admittance data set is augmented with white noise to simulate realistic measurement *** K-fold cross-validation technique is employed for training and testing the *** experimental validation of the proposed method shows that the proposed method can accurately identify the state and damage location in the plate with an average accuracy of 98%.Comparing with established 1D CNN models reveals superior performance of the proposed method,with significantly lower testing *** proposed method exhibits the ability to directly handle raw electromechanical admittance responses and extract optimal features,overcoming limitations associated with traditional piezoelectric admittance *** eliminating the need for signal preprocessing,this method holds promise for real-time damage monitoring of plate structures.

关键词： convolutional neural network electromechanical admittance electromechanical impedance piezoelectric transducer damage localization plate structure deep learning structural health monitoring

Enhancing Fire Detection Performance Based on Fine-Tuned YOLOv10

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第11期81卷 2281-2298页

作者： Trong Thua huynh Hoang thanh Nguyen Du Thang Phu Faculty of Information Technology 2 Posts and Telecommunications Institute of Technology(PTIT)Ho Chi Minh City700000Vietnam

In recent years,early detection and warning of fires have posed a significant challenge to environmental protection and human *** learning models such as Faster R-CNN(Faster Region based Convolutional Neural Network),YOLO(You Only Look Once),and their variants have demonstrated superiority in quickly detecting objects from images and videos,creating new opportunities to enhance automatic and efficient fire *** YOLO model,especially newer versions like YOLOv10,stands out for its fast processing capability,making it suitable for low-latency ***,when applied to real-world datasets,the accuracy of fire prediction is still not *** study improves the accuracy of YOLOv10 for real-time applications through model fine-tuning techniques and data *** core work of the research involves creating a diverse fire image dataset specifically suited for fire detection applications in buildings and factories,freezing the initial layers of the model to retain general features learned from the dataset by applying the Squeeze and Excitation attention mechanism and employing the Stochastic Gradient Descent(SGD)with a momentum optimization algorithm to enhance accuracy while ensuring real-time fire *** results demonstrate the effectiveness of the proposed fire prediction approach,where the YOLOv10 small model exhibits the best balance compared to other YOLO family models such as nano,medium,and ***,the study provides an experimental evaluation to highlight the effectiveness of model fine-tuning compared to the YOLOv10 baseline,YOLOv8 and Faster R-CNN based on two criteria:accuracy and prediction time.

关键词： Fire detection accuracy prediction time fine-tuning real-time YOLOv10 Faster R-CNN

Effects of NaOH Concentrations on Properties of the Thermal Power Plant Ashes-Bricks by Alkaline Activation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Wuhan University of Technology(Materials Science) 2020年第1期35卷 131-139页

作者： Ngo Si-Huy Le Thi thanh-Tam huynh Trong-Phuoc Department of Engineering and Technology Hong Duc UniversityNo.565Quang Trung StreetDong Ve WardThanh Hoa City 440000Viet Nam Department of Rural Technology College of Rural DevelopmentCan Tho UniversityCampus II3/2 StreetNinh Kieu DistrictCan Tho City 900000Viet Nam

The use of the thermal power plant ashes including fly ash(FA) and bottom ash(BA) for producing unfired building bricks(UBB) using sodium hydroxide(NaOH) solution as an alkaline activator was investigated. A low applied forming pressure of 0.5 MPa and various NaOH concentrations of 5, 8, 10, and 12 M were used for the preparation of brick samples with different solution-to-binder(S/B) ratios of 0.35 and 0.40. The bricks were subjected to various test programs with reflecting the effect of both NaOH concentrations and S/B ratios on the brick’s properties. The compressive strength, unit weight, ultrasonic pulse velocity, and thermal conductivity of bricks increased with increasing NaOH concentration, whereas the contrary trend was found with increasing S/B ratio. Also, the water absorption of bricks was observed to reduce with increasing NaOH concentration and decreasing S/B ratio. As the results, the combined utilization of both low forming pressure and coal power plant ashes can produce the UBBs with low unit weight, low heat conductivity, and acceptable strength and water absorption rate as stipulated by TCVN 6477-2016. Furthermore, the outcomes of chemical analysis and microstructure observation also demonstrate that a high concentration of the Na OH promoted the geopolymerization process. Notably, the use of NaOH solution of either 10 M or above is recommended for the production of UBBs, which are classified as grade M5.0 or higher.

关键词： unfired building brick fly ash bottom ash thermal power plant ashes sodium hydroxide solution low forming pressure

维普期刊数据库

Deep Learning Based Optimal Multimodal Fusion Framework for Intrusion Detection Systems for Healthcare Data

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2021年第3期66卷 2555-2571页

作者： Phong thanh Nguyen Vy Dang Bich huynh Khoa Dang Vo Phuong thanh Phan Mohamed Elhoseny Dac-Nhuong Le Department of Project Management Ho Chi Minh City Open UniversityHo Chi Minh City700000Vietnam Department of Learning Material Ho Chi Minh City Open UniversityHo Chi Minh City700000Vietnam Faculty of Computers and Information Mansoura UniversityDakahlia Governorate35516Egypt Institute of Research and Development Duy Tan UniversityDanang550000Vietnam Faculty of Information Technology Duy Tan UniversityDanang550000Vietnam

Data fusion is a multidisciplinary research area that involves different *** is used to attain minimum detection error probability and maximum reliability with the help of data retrieved from multiple healthcare *** generation of huge quantity of data from medical devices resulted in the formation of big data during which data fusion techniques become *** medical data is a crucial issue of exponentially-pacing computing world and can be achieved by Intrusion Detection Systems(IDS).In this regard,since singularmodality is not adequate to attain high detection rate,there is a need exists to merge diverse techniques using decision-based multimodal fusion *** this view,this research article presents a new multimodal fusion-based IDS to secure the healthcare data using *** proposed model involves decision-based fusion model which has different processes such as initialization,pre-processing,Feature Selection(FS)and multimodal classification for effective detection of *** FS process,a chaotic Butterfly Optimization(BO)algorithmcalled CBOA is *** the classic BO algorithm offers effective exploration,it fails in achieving faster *** order to overcome this,i.e.,to improve the convergence rate,this research work modifies the required parameters of BO algorithm using chaos ***,to detect intrusions,multimodal classifier is applied by incorporating three Deep Learning(DL)-based classification ***,the concepts like Hadoop MapReduce and Spark were also utilized in this study to achieve faster computation of big data in parallel computation *** validate the outcome of the presented model,a series of experimentations was performed using the benchmark NSLKDDCup99 Dataset *** proposed model demonstrated its effective results on the applied dataset by offering the maximum accuracy of 99.21%,precision of 98.93%and detection rate of 99.59%.The results assured the betterm

关键词： Big data data fusion deep learning intrusion detection bio-inspired algorithm spark

An Optimal Deep Learning Based Computer-Aided Diagnosis System for Diabetic Retinopathy

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2021年第3期66卷 2815-2830页

作者： Phong thanh Nguyen Vy Dang Bich huynh Khoa Dang Vo Phuong thanh Phan Eunmok Yang Gyanendra Prasad Joshi Department of Project Management Ho Chi Minh City Open UniversityHo Chi Minh City7000000Vietnam Department of Learning Material Ho Chi Minh City Open UniversityHo Chi Minh City7000000Vietnam Department of Convergence Science Kongju National UniversityGongju32588South Korea Department of Computer Science and Engineering Sejong UniversitySeoul05006South Korea

Diabetic Retinopathy(DR)is a significant blinding disease that poses serious threat to human vision *** and severity grading of DR are difficult processes to ***,it depends on ophthalmoscopically-visible symptoms of growing severity,which is then ranked in a stepwise scale from no retinopathy to various levels of DR *** paper presents an ensemble of Orthogonal Learning Particle Swarm Optimization(OPSO)algorithm-based Convolutional Neural Network(CNN)Model EOPSO-CNN in order to perform DR detection and *** proposed EOPSO-CNN model involves three main processes such as preprocessing,feature extraction,and *** proposed model initially involves preprocessing stage which removes the presence of noise in the input ***,the watershed algorithm is applied to segment the preprocessed *** by,feature extraction takes place by leveraging EOPSO-CNN ***,the extracted feature vectors are provided to a Decision Tree(DT)classifier to classify the DR *** study experiments were carried out using Messidor DR Dataset and the results showed an extraordinary performance by the proposed method over compared methods in a considerable *** simulation outcome offered the maximum classification with accuracy,sensitivity,and specificity values being 98.47%,96.43%,and 99.02%respectively.

关键词： Diabetic retinopathy convolutional neural network classification image processing computer-aided diagnosis

Novel Deletion in Exon 7 of Betaine Aldehyde Dehydrogenase 2(BADH2)

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Rice science 2023年第2期30卷 104-112,I0020-I0030页

作者： VAN Quoc Giang huynh Ky NGUYEN Chau thanh Tung NGUYEN Loc Hien NGUYEN van Manh NGUYEN Nhut thanh VO Cong thanh SWEE Keong Yeap College of Agriculture Can Tho UniversityCan Tho 92000Vietnam Department of Sciences and Technology of Long An Province Long An 850000Vietnam China-Association of Southeast Asian Nations College of Marine Sciences Xiamen University MalaysiaSepang 49000SelangorMalaysia

The fragrance of rice is one of the premium characteristics that breeders want to include in rice varieties due to the higher market value. Nucleotide deletions in exons 2(7 bp) and 7(8 bp) of Betaine Aldehyde Dehydrogenase 2(BADH2) are associated with fragrance in rice. In this study, a new 13 bp deletion in exon 7 of the BADH2 gene was discovered in the Nang Thom Cho Dao(NTCD) variety, and the mutation has been closely related to the genetic background of indica subspecies through the Bayesian phylogenetic approach and haplotype network analysis of the 3 000 Rice Genomes Project. In addition, a set of functional markers(EX07-13F, EX07-13RN, and EX07-13RM) identified the 13 bp deletion only within NTCD(no amplified band) compared with both non-aromatic and other aromatic rice varieties(110 bp band). The deletion of 13 bases instead of 8 bases in exon 7 of BADH2 caused a premature stop codon, which down-regulated the expression of the BADH2 transcript while associated with up-regulation of OsP5CS and the high amount of 2-acetyl-1-pyrroline. It is potential to use the deletion in exon 7 of the BADH2 gene as a novel marker for adulteration and breeding of fragrant rice varieties, particularly for NTCD.

关键词： novel deletion BADH2 fragrant rice functional marker

维普期刊数据库

Minimizing Warpage for Macro-Size Fused Deposition Modeling Parts

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2021年第9期68卷 2913-2923页

作者： thanh Thuong huynh Tien V.T.Nguyen Quoc Manh Nguyen Trieu Khoa Nguyen Department of Mechanical Engineering Can Tho UniversityCampus IICan ThoVietnam Faculty of Mechanical Technology Industrial University of Ho Chi Minh CityHo Chi Minh70000Vietnam Mechanical Engineering Department National Kaohsiung University of Science and TechnologyKaohsiung80778Taiwan Faculty of Mechanical Engineering Hung Yen University of Technology and EducationHung Yen16000Vietnam

In this study,we investigated warpage and corner lifting minimization for three-dimensional printed parts generated by macro-size fused deposition modeling(FDM).First,the reasons for warpage were theoretically *** approach revealed that the thermal deformation and differential volumetric shrinkage of the extruded molten plastic resulted in warpage of FDM *** addition,low adhesion between the deposited model and the heated or non-heated printing bed may intensify warpage *** a next step,initial small-size and medium-size models were used to identify parameters to manage and minimize warpage in a way that would reduce material consumption and running ***,a macro-size model was built to experimentally investigate and verify the technical solutions to minimize the warpage of FDM *** conclusion,an improved part with reduced warpage was efficiently produced after detailed consideration of thermal effects and adhesion *** exists to widen the application scope of FDM technology in manufacturing for processes like thermoforming that involve mold core fabrication with *** technology,which has applications not only in mechanical engineering but also in related engineering fields,is convenient and could readily be applied to practical manufacturing industries.

关键词： Fused deposition modeling corner lifting volumetric shrinkage thermal deformation acrylonitrile butadiene styrene filament

CONVERGENCE RESULTS FOR NON-OVERLAP SCHWARZ WAVEFORM RELAXATION ALGORITHM WITH CHANGING TRANSMISSION CONDITIONS

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Acta Mathematica Scientia 2022年第1期42卷 105-126页

作者： Minh-Phuong TRAN thanh-Nhan NGUYEN Phuoc-Toan huynh Nhu-Binh LY Minh-Dang NGUYEN Quoc-Anh HO Applied Analysis Research Group Faculty of Mathematics and StatisticsTon Duc Thang UniversityHo Chi Minh cityVietnam Department of Mathematics Ho Chi Minh City University of EducationHo Chi Minh cityVietnam Faculty of Mathematics and Statistics Ton Duc Thang UniversityHo Chi Minh cityVietnam

In this paper,we establish a new algorithm to the non-overlapping Schwarz domain decomposition methods with changing transmission conditions for solving one dimensional advection reaction diffusion *** precisely,we first describe the new algorithm and prove the convergence results under several natural assumptions on the sequences of parameters which determine the transmission *** we give a simple method to estimate the new value of parameters in each *** interesting advantage of our method is that one may update the better parameters in each iteration to save the computational cost for optimizing the parameters after many *** some numerical experiments are performed to show the behavior of the convergence rate for the new method.

关键词： domain decomposition method Schwarz waveform relaxation algorithm advection reaction diffusion changing transmission conditions

维普期刊数据库