检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2025年第3期6卷 589-603页

作者： Qiu, Jian-Qiang Zhang, Chun-Yang Chen, C. L. Philip Fuzhou University College of Computer and Data Science Fuzhou350108 China South China University of Technology School of Computer Science and Engineering Guangzhou510006 China

Pretrained language models (PLMs) have shown remarkable performance on question answering (QA) tasks, but they usually require fine-tuning (FT) that depends on a substantial quantity of QA pairs. Therefore, improving the performance of PLMs in scenarios with only a small number of training examples, also known as a few-shot setting, is of great practical significance. Current mitigation strategies for the few-shot QA task largely rely on pretraining a QA task-specific language model from scratch, overlooking the potential of foundational PLMs to generate QA pairs, particularly in the few-shot setting. To address this issue, we propose a prompt-based QA data augmentation method aimed at automating the creation of high-quality QA pairs. It employs the PFT method, adapting the question generation process of PLMs to the few-shot setting. Additionally, we introduce a dynamic text filling training strategy. This strategy simulates the progressive human learning process, thereby alleviating overfitting of PLMs in the few-shot setting and enhancing their reasoning capability to tackle complex questions. Extensive experiments demonstrate that the proposed method outperforms existing approaches across various few-shot configurations. 2691-4581 © 2024 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See https://***/publications/rights/*** for more information.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Evaluating the Effectiveness of Medication Across Diseases 2

Evaluating the Effectiveness of Medication Across Diseases

引用

2nd International Conference on Computing and data Analytics, ICCDA 2024

作者： Miracline, A. Sheryl Hepsiba, C. Linda Bishop Heber College Department of Data Science Tiruchirappalli India

ISBN: (纸本)9798331542573

The method to improve healthcare outcomes and to treat all diseases, from acute bacterial sinusitis and adenocarcinoma of the pancreas, and all the way to atopic dermatitis, is through medication effectiveness assessment. This work develops a proposal for a patient review-based machine learning model that evaluates the efficacy of medications for 37 commonly treated medical conditions. A general shortcoming in drug performance algorithms is the fact that they often rely purely on clinical trials or opinions of drug experts, disregarding real user feedback within the world itself. While these have, to a great extent, proved to be very instrumental in the advancement of medical science, they always miss patient-centric factors such as ease of use and satisfaction-mostly critical for long-term effectiveness of medication and adherence. The study bridges some gaps with the analysis of a dataset including drug names, types, average price, reviews, and performance metrics such as effectiveness, ease of use, and satisfaction. Our model summarizes these data to forecast which drugs would best suit a given condition based on user experience. Our proposed model provides for an overall better medication evaluation by applying ML algorithms to user reviews. It effectively closes the gap between outcomes reported by patients and clinical efficacy. Such research would help prescribing officers provide the most efficacious medicine to patients, thereby reducing their workload and helping in better care of the patients. © 2024 IEEE.

关键词： Dermatitis

来源：评论

学校读者我要写书评

暂无评论

Application of SP-UNet++ in Remote Sensing Image Segmentation

Application of SP-UNet++ in Remote Sensing Image Segmentatio...

引用

2024 Cross Strait Radio science and Wireless Technology Conference, CSRSWTC 2024

作者： Li, Yifan Wu, Gengshen City University of Macau Faculty of Data Science Macau China

ISBN: (纸本)9798331507794

In the field of image segmentation, U-Net and Fully Convolutional Networks (FCNs) have achieved great success, but these models have two limitations, their optimal depth is unknown a priori, and requires extensive schema search or inefficient integration of different depth models;Their hop connections impose an unnecessarily restrictive fusion scheme that forces aggregation only on the same scale feature maps of the encoder and decoder subnetworks. It is more prominent in the field of remote sensing. In order to overcome these two problems, this paper further adds the spatial pyramid pooling module called Spatial Pyramid Pooling U-Net++ (SP-UNet++) based on U-Net++. Automatically learning features at different depths and determining the depth with the best performance can greatly reduce the number of training parameters. Experimental results show that the overall recognition accuracy of SP-UNet++ is significantly improved compared with U-Net. © 2024 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Adversarial Text Generation with Dynamic Contextual Perturbation

Adversarial Text Generation with Dynamic Contextual Perturba...

引用

2024 IEEE Calcutta Conference, CALCON 2024

作者： Waghela, Hetvi Sen, Jaydip Rakshit, Sneha Dasgupta, Subhasis Praxis Business School Dep of Data Science Kolkata India

ISBN: (纸本)9798331516666

Adversarial attacks on Natural Language Processing (NLP) models expose vulnerabilities by introducing subtle perturbations to input text, often leading to misclassification while maintaining human readability. Existing methods typically focus on word-level or local text segment alterations, overlooking the broader context, which results in detectable or semantically inconsistent perturbations. We propose a novel adversarial text attack scheme named Dynamic Contextual Perturbation (DCP). DCP dynamically generates context-aware perturbations across sentences, paragraphs, and documents, ensuring semantic fidelity and fluency. Leveraging the capabilities of pre-trained language models, DCP iteratively refines perturbations through an adversarial objective function that balances the dual objectives of inducing model misclassification and preserving the naturalness of the text. This comprehensive approach allows DCP to produce more sophisticated and effective adversarial examples that better mimic natural language patterns. Our experimental results, conducted on various NLP models and datasets, demonstrate the efficacy of DCP in challenging the robustness of state-of-the-art NLP systems. By integrating dynamic contextual analysis, DCP significantly enhances the subtlety and impact of adversarial attacks. This study highlights the critical role of context in adversarial attacks and lays the groundwork for creating more robust NLP systems capable of withstanding sophisticated adversarial strategies. © 2024 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Overlapping Tensor Train Completion with TV Regularization for Image Reconstruction 5

Overlapping Tensor Train Completion with TV Regularization f...

引用

5th International Conference on Computer, Big data and Artificial Intelligence, ICCBD+AI 2024

作者： Mo, Lang Gong, Wenwu Yang, Lili Southern University of Science and Technology Department of Statistics and Data Science Shenzhen China

ISBN: (纸本)9798331533991

In this paper, we propose a novel tensor completion framework, Overlapping Tensor Train Completion with TV Regularization (OTTC-TV), which integrates the strengths of both Overlapping Ket Augmentation (OKA) and Total Variation (TV) regularization. This framework addresses two critical limitations of traditional tensor completion methods based on the Tensor Train (TT) decomposition: poor recovery performance in high missing rate scenarios and block artifacts that disrupt image continuity. By leveraging OKA, our proposal reduces block artifacts and improves inter-block smoothness, overcoming the non-smooth transitions between blocks. TV regularization further ensures smoothness across the reconstructed tensor by penalizing abrupt gradient changes. The proposed model demonstrates superior performance in completing high-missing-rate color images while also providing greater flexibility in handling varying input tensor dimensions. Extensive experiments validate the effectiveness and adaptability of OTTC-TV in color image reconstruction tasks. © 2024 IEEE.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Advanced Parkinson’s Disease Classification using Hybrid Model: SVM and Random Forest 3

Advanced Parkinson’s Disease Classification using Hybrid Mo...

引用

3rd International Conference on Intelligent data Communication Technologies and Internet of Things, IDCIoT 2025

作者： Archana, S.H. Rajeshkannan, M. Anish, M.J. Sandeep, Pasupuleti Artificial Intelligence and Data Science Saveetha Engineering College Chennai India

ISBN: (纸本)9798331527549

Parkinson’s disease (PD) is a progressive and chronic neurodegenerative condition that significantly impairs motor and cognitive functions, posing a considerable burden on global healthcare systems. Early and precise diagnosis is vital to improving patient outcomes, yet traditional diagnostic techniques often encounter limitations due to overlapping symptoms with other neurological disorders and the subtle nature of early PD indicators. This research introduces a machine learning-driven approach that employs a hybrid ensemble model integrating Support Vector Machine (SVM) and Random Forest classifiers to enhance diagnostic accuracy by analyzing vocal and acoustic features. To address dataset imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) is applied, enabling the model to identify meaningful patterns within underrepresented data. Principal Component Analysis (PCA) is utilized to reduce dimensionality, preserving critical data variance while minimizing computational complexity and mitigating overfitting. Hyperparameter tuning further optimizes each classifier, bolstering the model’s accuracy and robustness. Comparative analyses demonstrate that this hybrid ensemble model effectively distinguishes PD patients from healthy individuals, showcasing its potential as a reliable predictive tool in clinical applications. This study highlights the promise of advanced machine learning methodologies in supporting early PD detection, offering clinicians a valuable resource for timely intervention and improved patient care. © 2025 IEEE.

关键词： hybrid classification Machine learning models Parkinson’s Disease principal component analysis (PCA) random forest algorithm support vector machine (SVM)

来源：评论

学校读者我要写书评

暂无评论

Adversarial Robustness Through Dynamic Ensemble Learning

Adversarial Robustness Through Dynamic Ensemble Learning

引用

2024 IEEE Silchar Subsection Annual Conference, SILCON 2024

作者： Waghela, Hetvi Sen, Jaydip Rakshit, Sneha Dasgupta, Subhasis Praxis Business School Dept of Data Science Kolkata India

ISBN: (纸本)9798331540821

Adversarial attacks present considerable challenges to the reliability of widely used pretrained language models like GPT, BERT, ALBERT, and RoBERTa. This paper presents Adversarial Robustness through Dynamic Ensemble Learning (ARDEL), a novel scheme designed to enhance the robustness of PLMs against such attacks. ARDEL incorporates Randomized Smoothing and Masked Inference (RSMI) to create a smoothed classifier and increase resistance to adversarial attacks. Leveraging the diversity of multiple PLMs, ARDEL dynamically adjusts the ensemble configuration based on input characteristics and detected adversarial patterns. Key components include a meta-model for dynamic weighting, an adversarial pattern detection module, and adversarial training with regularization techniques. Comprehensive evaluations using standardized datasets and various adversarial attack scenarios demonstrate that ARDEL, especially with RSMI and ensemble methods, significantly improves robustness compared to existing methods. By dynamically reconfiguring the ensemble to prioritize the most robust models for each input, ARDEL effectively reduces attack success rates and maintains higher accuracy under adversarial conditions. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Multilingual Sentiment Analysis for Investigating Perceptions of Globalization 24

Multilingual Sentiment Analysis for Investigating Perception...

引用

24th IEEE International Conference on data Mining Workshops, ICDMW 2024

作者： Ani, Anagha Cambria, Erik Nanyang Technological University College of Computing & Data Science Singapore

ISBN: (纸本)9798331530631

This paper aims to identify trends in perceptions on the topic of globalization by performing sentiment analysis on text data collected from social media posts. A novel methodology of extracting culture-informed sentiments is proposed and tested on a corpus containing text posts from the social media site Reddit in two languages: French and English. To do so, a graph convolutional network is used to train a polarity classification model by extracting commonsense culture-specific knowledge using the SenticNet knowledge base. A variety of sentiment analysis tasks including polarity classification, intensity ranking, toxicity spotting, wellbeing assessment, and personality classification are performed using Sentic APIs on two extracted subsets of the corpus and the resulting trends in the data are identified and analyzed. © 2024 IEEE.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

SStyleGAN: a StyleGAN model for generating symmetrical lace images

SStyleGAN: a StyleGAN model for generating symmetrical lace ...

引用

2023 International Conference on Computer Graphics, Artificial Intelligence, and data Processing, ICCAID 2023

作者： Li, Jian Zhong, Shangping Chen, Kaizhi College of Computer and Data Science Fuzhou University Fuzhou China

ISBN: (纸本)9781510677685

Lace texture, as a manually designed texture image, needs to possess a series of essential aesthetic characteristics, such as periodicity, symmetry, and blank-leaving in artistic design creation. It requires human designers to spend a lot of time and effort, so it is necessary to apply generative models to generate lace images. In image generation tasks, compared to models such as DCGAN, CycleGAN, and ProGAN, although images generated using StyleGAN2 perform well in terms of resolution, texture details, and periodicity, they still perform poorly in terms of symmetry in lace images. To address the above issues, this article proposes an improved model SStyleGAN (Symmetry StyleGAN) based on StyleGAN2. In terms of discriminators, in order to enhance the attention of the proposed model to image symmetry, we have added a symmetry discriminator, that is, SStyleGAN adopts a dual discriminator structure;In terms of generator, in order to improve the similarity of the feature maps on the left and right sides of the lace image, this paper adds a mean square error loss term based on the loss function of StyleGAN2;In terms of noise input, in order to control the symmetry of the lace image at details such as lace edges, the noise of the StyleGAN2 model is modified to a symmetrical structure, so that the noise input itself has symmetry. In addition to the commonly used FID (Fréchet Insertion Distance) in the generative model, we also used the SSIM (Structural Similarity) metric for the evaluation of the experimental results in this article to detect the symmetry of the generated images. The experimental results show that compared to the lace images generated by the StyleGAN2 model, the lace images generated by the model proposed in this paper not only inherit the advantages of the former, but also have symmetry characteristics. © 2024 SPIE.

关键词： Discriminators

来源：评论

学校读者我要写书评

暂无评论

Research on Game Recommendation Algorithms Based on Hybrid Models 24

Research on Game Recommendation Algorithms Based on Hybrid M...

引用

3rd International Conference on Computer, Artificial Intelligence and Control Engineering, CAICE 2024

作者： Wang, Yonghao Chu, Chiawei Li, Kaiye Faculty of Data Science City University of Macau Macau China

ISBN: (纸本)9798400716942

Game recommendation algorithms can help users narrow down their search and find games of interest. The mainstream game recommendation algorithms currently are collaborative filtering game recommendation algorithms. However, collaborative filtering mainly relies on user ratings and similarity to make recommendations. When user data includes a substantial amount of social information, this method fails to analyze the users' social data comprehensively. It cannot grasp the users' attitudes, opinions, and emotions, leading to inaccurate recommendation results. In this paper, we propose a hybrid recommendation model. In this model, we combine collaborative filtering algorithms with sentiment analysis, taking both users' gaming data and social data into account within the hybrid model. After selecting the most appropriate K value, we use the KNN classification algorithm for personalized recommendations, optimizing the user-based collaborative filtering game recommendation algorithm. The evaluation of the recommendation results is derived from MSE, Accuracy, F1, Precision, and Recall. The results indicate that the recommendation algorithm, after hybridization, shows an improvement in recommendation accuracy compared to the user-based collaborative filtering game recommendation algorithm, proving the effectiveness of this research. © 2024 ACM.

关键词： Collaborative filtering

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：