检索结果-内蒙古大学图书馆

KeyEE:Enhancing Low-Resource Generative Event Extraction with Auxiliary Keyword Sub-Prompt

big data Mining and Analytics 2024年第2期7卷 547-560页

作者： Junwen Duan Xincheng Liao Ying An Jianxin Wang Hunan Key Laboratory of Bioinformatics School of Computer Science and EngineeringCentral South UniversityChangsha 410083China Big Data Institute Central South UniversityChangsha 410083China

Event Extraction(EE)is a key task in information extraction,which requires high-quality annotated data that are often costly to *** classification-based methods suffer from low-resource scenarios due to the lack of label semantics and fine-grained *** recent approaches have endeavored to address EE through a more data-efficient generative process,they often overlook event keywords,which are vital for *** tackle these challenges,we introduce KeyEE,a multi-prompt learning strategy that improves low-resource event extraction by Event Keywords Extraction(EKE).We suggest employing an auxiliary EKE sub-prompt and concurrently training both EE and EKE with a shared pre-trained language *** the auxiliary sub-prompt,KeyEE learns event keywords knowledge implicitly,thereby reducing the dependence on annotated ***,we investigate and analyze various EKE sub-prompt strategies to encourage further research in this *** experiments on benchmark datasets ACE2005 and ERE show that KeyEE achieves significant improvement in low-resource settings and sets new state-of-the-art results.

关键词： natural language processing Event Extraction(EE) Multi-Prompt Learning(MPL) low-resource

来源：评论

学校读者我要写书评

暂无评论

Efect of Pre‑stretch Strain at High Temperatures on the Formability of AZ31 Magnesium Alloy Sheets

引用

Acta Metallurgica Sinica(English Letters) 2023年第1期36卷 48-60页

作者： Yongqiao Li Lifei Wang Xiaohuan Pan Qiang Zhang Guangsheng Huang Bin Xing Weili Cheng Hongxia Wang Kwang Seon Shin College of Materials Science and Engineering Taiyuan University of TechnologyTaiyuan 030024China Chongqing Innovation Center of Industrial Big-Data Co.Ltd. National Engineering Laboratory for Industrial Big-Data Application TechnologyChongqing 400707China School of Materials Science and Engineering Shanghai Jiao Tong UniversityShanghai 200240China National Engineering Research Center for Magnesium Alloys Chongqing UniversityChongqing 400044China Magnesium Technology Innovation Center School of Materials Science and EngineeringSeoul National UniversitySeoul 881416Republic of Korea

High-temperature pre-stretching experiments were carried out on the AZ31 Mg alloy at 723 K with strain levels of 2.54%,6.48%,10.92%,and 19.2%to alter the microstructure and texture for improving room-temperature *** results showed that the strain-hardening coefcient increased,while the Lankford value *** addition,the Erichsen values of all pre-stretch sheets were enhanced compared with that of the as-received *** maximum Erichsen value increased from 2.38 mm for the as-received sample to 4.03 mm for the 10.92%-stretched sample,corresponding to an improvement of 69.32%.This improvement was mainly attributed to the gradual increase in grain size,and the(0001)basal texture was weakened due to the activated non-basal slip as the high-temperature pre-stretching strain levels *** visco-plastic self-consistent analysis was performed on the as-received and high-temperature pre-stretched *** confrmed the higher activity of the prismatic slip in 10.92%-stretched sample,leading to divergence and weakening of basal texture *** results in an augmentation of the Schmid factor under diferent slip ***,it can be concluded that high-temperature pre-stretching technology provided an efective method to enhance the formability of Mg alloy sheets.

关键词： High-temperature deformation Grain growth Magnesium alloys Texture Visco-plastic self-consistent modeling Formability

来源：评论

学校读者我要写书评

暂无评论

Federated Learning Model for Auto Insurance Rate Setting Based on Tweedie Distribution

引用

Computer Modeling in Engineering & sciences 2024年第1期138卷 827-843页

作者： Tao Yin Changgen Peng Weijie Tan Dequan Xu Hanlin Tang State Key Laboratory of Public Big Data Guizhou UniversityGuiyang550025China Guizhou Big Data Academy Guizhou UniversityGuiyang550025China Key Laboratory of Advanced Manufacturing Technology Ministry of EducationGuizhou UniversityGuiyang550025China College of Computer Science and Technology Guizhou UniversityGuiyang550025China ChinaDataPay Company Guiyang550025China

In the assessment of car insurance claims,the claim rate for car insurance presents a highly skewed probability distribution,which is typically modeled using Tweedie *** traditional approach to obtaining the Tweedie regression model involves training on a centralized dataset,when the data is provided by multiple parties,training a privacy-preserving Tweedie regression model without exchanging raw data becomes a *** address this issue,this study introduces a novel vertical federated learning-based Tweedie regression algorithm for multi-party auto insurance rate setting in data *** algorithm can keep sensitive data locally and uses privacy-preserving techniques to achieve intersection operations between the two parties holding the *** determining which entities are shared,the participants train the model locally using the shared entity data to obtain the local generalized linear model intermediate *** homomorphic encryption algorithms are introduced to interact with and update the model intermediate parameters to collaboratively complete the joint training of the car insurance rate-setting *** tests on two publicly available datasets show that the proposed federated Tweedie regression algorithm can effectively generate Tweedie regression models that leverage the value of data fromboth partieswithout exchanging *** assessment results of the scheme approach those of the Tweedie regressionmodel learned fromcentralized data,and outperformthe Tweedie regressionmodel learned independently by a single party.

关键词： Rate setting Tweedie distribution generalized linear models federated learning homomorphic encryption

来源：评论

学校读者我要写书评

暂无评论

Wordle data Exploration and Prediction Based on BP-GA Model 4

Wordle Data Exploration and Prediction Based on BP-GA Model

引用

4th International Conference on Computers, Information Processing and Advanced Education, CIPAE 2023

作者： Zhou, Luyao Ruan, Peiguang Wenzhou University of Technology School of Data Science and Artificial Intelligence Wenzhou China Zhuhai College of Science and Technology Alibaba Cloud Big Data Application College Zhuhai China

ISBN: (纸本)9798350342710

Wordle, a fun puzzle game, the acquisition of Wordle by the New York Times led to a surge in the game's popularity. In this article, we aim to explore various data related to Wordle, including the number of daily reported results, words of varying difficulty, and the distribution of user attempts. We use the ARIMA model to predict the number of reported results in the world in the short term, and verify it with the LSTM model, and the results show that our interval prediction is correct. We forecast to reach 14,737 within the forecast range (13,667, 15,807) in the coming months. In addition, we built a BP-GA model to predict the distribution of future reported outcomes and found a number of word attributes, including seven numerical factors and two potential uncertainties that could affect reported outcomes. Based on the random forest model and grid search, the difficulty classification model of words is constructed, and it is found that word frequency and ASCII mean have great influence on the difficulty classification of words. The classification accuracy is 88.89%, which verifies the superiority of the model. © 2023 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Addition of the T5 exonuclease increases the prime editing efficiency in plants

引用

Journal of Genetics and Genomics 2023年第8期50卷 582-588页

作者： Zhen Liang Yuqing Wu Yingjie Guo Sha Wei School of Life Science Shanxi UniversityTaiyuanShanxi 030006China Research Institute of Big Data Science and Industry Shanxi UniversityTaiyuanShanxi 030006China

Prime editing(PE)is a versatile genome editing tool without the need for double-stranded DNA breaks or donor DNA templates,but is limited by low editing *** previously fused the M-MLV reverse transcriptase to the Cas9 nickase,generating the PE2(v1)system,but the editing efficiency of this system is still *** we develop different versions of PE2 by adding the 50-to-30 exonuclease at different positions of the nCas9-M-MLV RT fusion ***2(v2),in which the T5 exonuclease fused to the N-terminus of the nCas9-MMLV fusion protein enhances prime editing efficiency of base substitutions,deletions,and insertions at several genomic sites by 1.7-to 2.9-fold in plant cells compared to PE2(v1).The improved editing efficiency of PE2(v2)is further confirmed by generating increased heritable prime edits in stable transgenic plants compared to the previously established PE-P1,PE-P2,and PPE *** PE2(v2),we generate herbicide-resistant rice by simultaneously introducing mutations causing amino acid substitutions at two target *** PE efficiency is further improved by combining PE2(v2)and *** together,the increased genome editing efficiency of PE2(v2)developed in this study may enhance the applications of PE in plants.

关键词： Prime editing CRISPR T5 exonuclease Genome editing

来源：评论

学校读者我要写书评

暂无评论

Effectiveness of a touch-based hand therapeutic application for hand dexterity in children with handwriting difficulties-a pilot study

引用

Multimedia Tools and Applications 2024年 1-35页

作者： Smitha, John Renumol, V.G. School of Computer Science and Technology Division of Data Science and Cyber Security Karunya Institute of Technology and Sciences Tamil Nadu Coimbatore India School of Engineering Cochin University of Science and Technology Kochi India

Handwriting is an important skill for children during their academic years. It is the coordination of perceptual-motor and cognitive abilities. Some children have difficulties in handwriting, which is an indication of issues in their fine motor skills. Nowadays, technology-based instructions are available to develop fine motor skills in children with handwriting disabilities. This paper describes the details of a newly developed software application namely ‘HanDex’ for children with handwriting difficulties and the experiments conducted to test its effectiveness. The application comprises six hand therapeutic activities which were designed based on the oscillation theory of handwriting. Initially, in the first phase, the experiments were conducted in nine primary school children from various schools, aged 8–12 years. Then, in the second phase, the software training was given to six upper primary school children (10–12 years) with handwriting issues. The results of using this application were promising. So one more study was conducted in another set of nine children (8–11 years) with handwriting difficulties. The participants were assessed by their special educators, class teachers and parents. After the pretest, they received a total of 18 handwriting practice sessions using this new application. At the end of the training sessions, a post test was conducted, to measure the progress in handwriting. The handwriting rubrics with eight handwriting parameters such as letter formation, placement, letter sizing, spacing of letters, legibility, speed, neatness, and spacing of words were used for handwriting assessment. The findings from the study suggested that touch-based hand therapeutic activities were beneficial for children with writing difficulties to improve their handwriting performance and tripod grip. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Children with handwriting difficulties Hand Dexterity Handwriting performance and tripod grip Tablet computer ‘HanDex’ Software

来源：评论

学校读者我要写书评

暂无评论

Application of improved virtual sample and sparse representation in face recognition

引用

CAAI Transactions on Intelligence technology 2023年第4期8卷 1391-1402页

作者： Yongjun Zhang Zewei Wang Xuexue Zhang Zhongwei Cui Bob Zhang Jinrong Cui Lamin LJanneh College of Computer Science and Technology Guizhou UniversityGuiyangChina Big Data Science and Intelligent Engineering Research Institute Guizhou Education UniversityGuiyangChina School of Mathematics and Big Data Guizhou Education UniversityGuiyangChina Department of Computer and Information Science University of MacaoAvenida da UniversidadeTaipaMacaoChina School of Mathematics and Information South China Agricultural UniversityGuangzhouChina

Sparse representation plays an important role in the research of face *** a deformable sample classification task,face recognition is often used to test the performance of classification *** face recognition,differences in expression,angle,posture,and lighting conditions have become key factors that affect recognition ***,there may be significant differences between different image samples of the same face,which makes image classification very ***,how to build a robust virtual image representation becomes a vital *** solve the above problems,this paper proposes a novel image classification ***,to better retain the global features and contour information of the original sample,the algorithm uses an improved non‐linear image representation method to highlight the low‐intensity and high‐intensity pixels of the original training sample,thus generating a virtual ***,by the principle of sparse representation,the linear expression coefficients of the original sample and the virtual sample can be calculated,*** obtaining these two types of coefficients,calculate the distances between the original sample and the test sample and the distance between the virtual sample and the test *** two distances are converted into distance ***,a simple and effective weight fusion scheme is adopted to fuse the classification scores of the original image and the virtual *** fused score will determine the final classification *** experimental results show that the proposed method outperforms other typical sparse representation classification methods.

关键词： representation sample image

来源：评论

学校读者我要写书评

暂无评论

Complementary-Label Adversarial Domain Adaptation Fault Diagnosis Network under Time-Varying Rotational Speed and Weakly-Supervised Conditions

引用

Computers, Materials & Continua 2024年第4期79卷 761-777页

作者： Siyuan Liu Jinying Huang Jiancheng Ma Licheng Jing Yuxuan Wang School of Data Science and Technology North University of ChinaTaiyuan030051China School of Mechanical Engineering North University of ChinaTaiyuan030051China

Recent research in cross-domain intelligence fault diagnosis of machinery still has some problems,such as relatively ideal speed conditions and sample *** engineering practice,the rotational speed of the machine is often transient and time-varying,which makes the sample annotation increasingly ***,the number of samples collected from different health states is often *** deal with the above challenges,a complementary-label(CL)adversarial domain adaptation fault diagnosis network(CLADAN)is proposed under time-varying rotational speed and weakly-supervised *** the weakly supervised learning condition,machine prior information is used for sample annotation via cost-friendly complementary label learning.A diagnosticmodel learning strategywith discretized category probabilities is designed to avoidmulti-peak distribution of prediction *** adversarial training process,we developed virtual adversarial regularization(VAR)strategy,which further enhances the robustness of the model by adding adversarial perturbations in the target *** experiments on two case studies validated the superior performance of the proposed method.

关键词： Time-varying rotational speed weakly-supervised fault diagnosis domain adaptation

来源：评论

学校读者我要写书评

暂无评论

RPID: Boosting Transferability of Adversarial Attacks on Vision Transformers

RPID: Boosting Transferability of Adversarial Attacks on Vis...

引用

2024 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2024

作者： Wang, Shujuan Wang, Panpan Sun, Yunpeng Lian, Zhichao Li, Shuohao School of Cyberspace Security Nanjing University of Science and Technology Wuxi China National University of Defense Technology Laboratory for Big Data and Decision China

ISBN: (纸本)9781665410205

Vision Transformers (ViTs) have achieved excellent performance on many computer vision tasks, which has attracted attention of many researchers for their adversarial robustness. As a kind of black-box attack, transfer-based at-tacks usually use adversarial examples generated by a surrogate model to attack structurally different models. It is practical and poses a certain threat to the application of ViTs in critical security areas. Existing transfer-based attacks against ViTs suffer from weak adversarial transferability and noticeable perceptibility. In this work, we propose a method called Reduce Regional Perturbation Interaction and Differentiated (RPID) attack, which employs two strategies of reducing correlation between regional perturbations and adding differentiated perturbations to produce adversarial examples. Extensive experiments demonstrate that our proposed method improves the transferability of the baseline methods for adversarial attacks against ViTs while maintaining stealthiness. © 2024 IEEE.

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

Clustered Federated Multi-Task Learning: A Communication-and-Computation Efficient Sparse Sharing Approach

引用

IEEE Transactions on Wireless Communications 2025年第6期24卷 4824-4838页

作者： Ai, Yuhan Chen, Qimei Zhu, Guangxu Wen, Dingzhu Jiang, Hao Zeng, Jun Li, Ming Wuhan University School of Electronic Information Wuhan430072 China Shenzhen Research Institute of Big Data Shenzhen518172 China ShanghaiTech University School of Information Science and Technology Shanghai China Fiberhome Communication Technology Co. Ltd Hube Wuhan China

Federated multi-task learning (FMTL) is a promising technology to tackle one of the most severe non-independent and identically distributed (non-IID) data challenge in federated learning (FL), which treats each client as a single task and learns personalized models by exploiting task correlations. However, the transmission of individual task models generally results in a significant amount of communication overhead compared with global model broadcasting. Furthermore, related works mainly focus on FMTLs with default and static relationships among tasks, which obliterates the non-IID data characteristic. To address these issues, we propose a novel Clustered FMTL mechanism via Sparse Sharing (FedSS). Specifically, we introduce an iterative model pruning approach that trains customized client models to deal with the non-IID issue. Thereafter, we divide clients into different tasks according to their model similarities to promote communication efficiency. Based on clustered tasks, we introduce a sparse sharing mechanism that allows clients to share model parameters dynamically among different tasks to further boost the training performance. On the other aspect, the infertile communication resources would degrade the FMTL performance by restricting the personalized model transmissions. Hence, we first theoretically analyze the convergence performance of the proposed FedSS, which quantitatively unveils the relationship between the local model training performance and communication resources. Thereafter, we formulate a communication-and-computation efficient optimization problem via a joint sparsity ratio assignment and bandwidth allocation strategy. Closed-form expressions for the optimal sparsity ratio and bandwidth allocation are derived based on Lyapunov optimization and block coordinate update (BCU) algorithms. Numerical results illustrate that the proposed FedSS outperforms the benchmarks, and achieves an efficient communication and computation performance. © 2025 IE

关键词： Bandwidth

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：