检索结果-内蒙古大学图书馆

3rd International Conference on Neural Networks, Information and Communication engineering, NNICE 2023

作者： Chen, Yanming Lou, Shubin Shuai, Mingrui He, Kan An, Zhulin School of Computer Science and Technology Anhui University Hefei China School of Life Sciences Anhui University Hefei China Institute of Computing Technology Chinese Academy of Sciences Beijing China

ISBN: (纸本)9798350335972

Alzheimer's disease (AD) is a cognitive disorder that often occurs in the elderly. In the early stage of the disease, memory loss is the primary manifestation, which is clinically called mild cognitive impairment (MCI). At present, there is no way to prevent or reverse AD. Therefore, it is essential to diagnose the disease in time. There are many studies using computers to diagnose AD automatically. Because the cause of the disease is unclear, using deep learning to diagnose AD from MRI data is a hot research topic today. In addition, for the 3-D medical imaging field, it is challenging to build a large dataset for a particular disease. Contrastive learning demonstrates strong capabilities and provides a way to learn knowledge from the images themselves. However, in this process, the additional information of participants is not used, and many studies have found that factors such as sex and age significantly influence the disease. So we improved the existing training method to take the age and sex of the participants into account and improve the encoding ability of the neural network. We validate performance on the ADNI dataset;our method outperforms training from scratch and state-of-the-art methods on AD diagnosis tasks after fine-tuning. © 2023 IEEE.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

CERG-KBQA: Advancing KBQA-Systems through Classification, Enumeration, Ranking, and Generation Strategies 10

CERG-KBQA: Advancing KBQA-Systems through Classification, En...

引用

10th IEEE Smart World Congress, SWC 2024

作者： Chen, Junyu Gao, Hui Wang, YuHang Wu, XuCheng University of Electronic Science and Technology of China School of Computer Science and Engineering Chengdu China Kash Institute of Electronics and Information Industry Xinjiang China

ISBN: (纸本)9798331520861

Knowledge Base Question Answering (KBQA) systems are critical for retrieving accurate information from structured data in response to natural language queries. Existing approaches, however, often fall short in coverage and generalization, especially in compositional and zero-shot situations. To mitigate these issues, we introduce CERG-KBQA, a novel framework that employs classification, enumeration, ranking, and generation strategies to enhance logical form candidate coverage and system generalization. We provide a suite of diverse logical form templates designed to accommodate various question types, effectively improving the coverage of logical form candidates. To enhance generalization capability, shuffled candidate augmentation is proposed to prevent positional bias and promote a more equitable consideration of candidates. Our framework is thoroughly tested on the GRAILQA and WEBQSP datasets, demonstrating significant improvements over baseline models, particularly in compositional generalization. With an F1 score of 77.0 on GRAILQA and 78.4 on WEBQSP, CERG-KBQA's robust performance solidifies its effectiveness across diverse KBQA scenarios, providing a versatile solution for complex KBQA tasks. © 2024 IEEE.

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

Optimized Fabric Imperfection Detection via Image Analysis and Isolation Forest Technique

Optimized Fabric Imperfection Detection via Image Analysis a...

引用

2024 International Conference on Signal Processing, Computation, Electronics, Power and Telecommunication, IConSCEPT 2024

作者： Jamunadevi, C. Sathishkumar, V.E. Prasath, S. Sadhanayaki, S. Jaikishore, N. Kongu Engineering College Department of Computer Technology TamilNadu Erode Perundurai India School of Engineering and Technology Sunway University Jalan University Selangor 47500 Malaysia VET Institute of Arts and Science College Department of Computer Science Erode India Kongu Arts And Science College Department Of Computer Technology And Information Technology India Kongu Engineering College Perundurai Department of Computer Technology Erode India

ISBN: (纸本)9798331540685

Fabric defect detection plays a crucial role in ensuring the quality of textile products. Traditional methods often rely on manual inspection, which is time-consuming and prone to human error. In this research, a novel method for fabric defect detection based on machine learning technique for anomaly detection the Isolation Forest algorithm is presented. The Isolation Forest algorithm builds a random forest of isolation trees to take advantage of the idea of isolating anomalies in a dataset. For the purpose of identifying fabric defects, every fabric sample is represented by a collection of features, including structural attributes, colour, and texture. By separating the majority of non-defective samples from the defective fabric samples, the Isolation Forest algorithm effectively finds the defective fabric samples. Based on the Standard Fabric Defect, our model also obtains a true positive rate value of 0.95 without generating any false alarms on our dataset. © 2024 IEEE.

关键词： BIRCH (balanced iterative reducing and clustering using hierarchies) Algorithm Fabric Defect Detection Isolation Forest

来源：评论

学校读者我要写书评

暂无评论

Channel Equalizer Research Based on Deep Neural Networks 4

Channel Equalizer Research Based on Deep Neural Networks

引用

4th IEEE International Conference on Data science and computer Application, ICDSCA 2024

作者： Yang, Jingjing Wu, Tao Han, Rengang School of Information Science and Engineering East China University of Science and Technology Shanghai China College of Electronic Engineering Xi 'an University of Posts Telecommunications Xi’an China School of Architecture & Urban Planning Anhui Jianzhu University Hefei China

ISBN: (纸本)9798350368239

In digital communication, data transmission is impacted by channel distortion. Inter-Symbol Interference (ISI) arises from the dispersive characteristics of the channel. To enhance reliability and support high-speed communication, channel equalization techniques are employed at the receiver to mitigate the effects of ISI. This paper proposes an effective equalizer, named CLE, which combines Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks. As the terrestrial wireless channel is mainly affected by small-scale fading such as multipath fading and Doppler shift, the Rayleigh fading channel is modelled using the Jakes approach. Compared with traditional Decision-Feedback Equalizer (DFE), CLE can achieve the performance improvement of close to 4~6dB on the premise of reaching the same bit error rate. The experiment results show that CLE is more efficient and less complex than the equalizer with similar techniques, resulting in faster processing and reduced computational demands, making it more practical for real-world applications. It means that our method is better equipped to handle the fluctuations in channel conditions, maintaining reliable performance even as the environment changes. ©2024 IEEE.

关键词： Intersymbol interference

来源：评论

学校读者我要写书评

暂无评论

Spam Detection Using Firefly Optimization Technique 2

Spam Detection Using Firefly Optimization Technique

引用

2nd IEEE International Conference on IoT, Communication and Automation technology, ICICAT 2024

作者： Srivastav, Shashank Gupta, Alok Singh, Laxman Singh, Rajesh Kumar Tripathi, Kanak Dwivedi, Anshu Kumar Buddha Institute of Technology Department of Computer Science Engineering Gorakhpur India Babu Banarasi das Engineering College Department of Computer Science Engineering Lucknow India Bennett University School of Computer Science Engineering and Technology Greater Noida India KIPM College of Engineering & Technology Dept of CSE Gorakhpur India Galgotia College of Engineering & Technology Dept of MCA Greater Noida India

ISBN: (纸本)9798350368109

In this electronically era, communication using email has become an part of our lives. However, with the convenience of email comes the nuisance of spam messages flooding our inboxes. Detecting and filtering out these spam emails is crucial to maintain productivity and security in our online interactions. In this research project, we explore the performance of using the firefly algorithm, a nature-inspired optimization technique, for spam detection. By leveraging the unique behavior of fireflies in search and optimization, we aim to develop a robust and efficient spam detection system. We start by gathering a dataset containing a mix of spam and non-spam (ham) emails. After preprocessing the text data and extracting relevant features using techniques like Count-Vectorizer, we train a logistic regression model, a commonly used classifier for binary classification tasks. To enhance the performance of our model, we employ the Firefly algorithm to optimize its hyperactive parameters. This algorithm impersonates the flashing behavior of fireflies to iteratively search for the foremost combination of model parameters, maximizing its ability to distinguish between spam and legitimate emails. Our experimental results demonstrate promising outcomes, with the optimized model achieving high accuracy and performance measures such as precision, recall, and F1-score on a held-out test set. These findings highlight the potential of nature-inspired algorithms like the firefly algorithm to improve spam detection techniques and effectively combat email spam in real-world scenarios. © 2024 IEEE.

关键词： Logistic regression

来源：评论

学校读者我要写书评

暂无评论

Pindel-TD:A Tandem Duplication Detector Based on A Pattern Growth Approach

引用

Genomics, Proteomics & Bioinformatics 2024年第1期22卷 99-104页

作者： Xiaofei Yang Gaoyang Zheng Peng Jia Songbo Wang Kai Ye School of Computer Science and Technology Faculty of Electronic and Information EngineeringXi’an Jiaotong UniversityXi’an 710049China Center for Mathematical Medical the First Affiliated Hospital of Xi’an Jiaotong UniversityXi’an 710061China Genome Institute the First Affiliated Hospital of Xi’an Jiaotong UniversityXi’an 710061China MOE Key Lab for Intelligent Networks&Networks Security Faculty of Electronic and Information EngineeringXi’an Jiaotong UniversityXi’an 710049China School of Automation Science and Engineering Faculty of Electronic and Information EngineeringXi’an Jiaotong UniversityXi’an 710049China School of Life Science and Technology Xi’an Jiaotong UniversityXi’an 710049China Faculty of Science Leiden UniversityLeiden 2311 EZNetherland

Tandem duplication(TD)is a major type of structural variations(SVs)that plays an important role in novel gene formation and human ***,TDs are often missed or incorrectly classified as insertions by most modern SV detection methods due to the lack of specialized operation on TD-related mutational ***,we developed a TD detection module for the Pindel tool,referred to as Pindel-TD,based on a TD-specific pattern growth ***-TD is capable of detecting TDs with a wide size range at single nucleotide *** simulated and real read data from HG002,we demonstrated that Pindel-TD outperforms other leading methods in terms of precision,recall,F1-score,and ***,by applying Pindel-TD to data generated from the K562 cancer cell line,we identified a TD located at the seventh exon of SAGE1,providing an explanation for its high ***-TD is available for non-commercial use at https://***/xjtu-omics/pindel.

关键词： Tandem duplication Pattern growth Short-read sequencing Structural variation SAGE1

来源：评论

学校读者我要写书评

暂无评论

Investigating Labeled Cyberbullying Incidents on the Weibo Social Network 21

Investigating Labeled Cyberbullying Incidents on the Weibo S...

引用

21st International Conference on Networking, Sensing and Control, ICNSC 2024

作者： Xu, Jun Yu, Songnan Lu, Xiaoyu Sean Wu, Di Kang, Qi School of Cyber Science and Engineering Nanjing University of Science and Technology Nanjing210000 China New Jersey Institute of Technology Department of Electrical and Computer Engineering NewarkNJ United States

ISBN: (纸本)9798350365221

This study aims to investigate cyberbullying incidents on Weibo by constructing a comprehensive dataset of labeled conversations. We collect 89K social media sessions from 10K user profiles and manually annotated the data to identify instances of cyberbullying. We analyze cyberbullying based on three fundamental characteristics, examining its distribution across different topics and the relationship between posting times and cyberbullying incidents. Additionally, we explore the influence of user popularity on the similarity of receiving bullying comments. Our study provides valuable insights into the characteristics of cyberbullying on Chinese social media and highlights the need for culturally sensitive detection models. © 2024 IEEE.

关键词： Phishing

来源：评论

学校读者我要写书评

暂无评论

NetPrompt: Neural Network Prompting Enhances Event Extraction in Large Language Models

引用

IEEE Transactions on Big Data 2025年

作者： Mu, Lin Cheng, Yide Shen, Jun Zhang, Yiwen Zhong, Hong School of Computer Science and Technology Anhui University Anhui Hefei230601 China The Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education the Anhui Engineering Laboratory of IoT Security Technologies the School of Computer Science and Technology the Institute of Physical Science and Information Technology Anhui University Hefei230039 China

Event Extraction involves extracting event-related information such as event types and event arguments from context, which has long been tackled through well-designed neural networks or fine-tuned pre-trained language models. These approaches require substantial annotated data for tuning parameters and are resource-intensive. Recently, Prompting strategies with frozen parameters, such as Chain-of-Thought and Self-Consistency, have delivered success in NLP using LLMs by generating intermediate thought steps. However, they suffer from the challenge of error propagation and lack of interaction between different thoughts. In this paper, we propose Neural Network-based Prompting (NetPrompt), a novel network-structured prompting strategy for event extraction. The core idea behind NetPrompt is to imitate the excellent information integration capabilities of neural network structures. Specifically, we first decompose the event extraction problem into diverse intermediate subtasks, and each subtask is represented as a node in different layers of the network, the output of the nodes in the preceding layer is fed into the subsequent layer. Secondly, we propose pruning strategies to adapt the reasoning overhead to different problems. Finally, we have conducted extensive experiments on two widely used event extraction benchmarks to evaluate NetPrompt. The results demonstrated that NetPrompt significantly improved the event extraction performance compared to previous methods. ©2015 IEEE.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

Developing a Smart Drip Infusion System to Elevate Precision in Healthcare 4

Developing a Smart Drip Infusion System to Elevate Precision...

引用

4th International Conference on Ubiquitous Computing and Intelligent Information Systems, ICUIS 2024

作者： Molefinyane, Sefemo Yahya, Abid Hamsadhwani, V. Rohini, R. Samikannu, Ravi Begam, K.Maruliya Botswana International University of Science and Technology Department of Electrical and Communications Systems Engineering Palapye Botswana Periyar Maniammai Institute of Science and Technology Department of Electrical and Electronics Engineering Thanjavur India Vivekanandha College of Engineering for Women Department of Computer Science and Engineering Tiruchengode India Saveetha School of Engineering Saveetha Institute of Medical and Technology Science Department of Computer Science and Communications Engineering Tiruchengode India

ISBN: (纸本)9798331529635

Intravenous (IV) therapy is a critical component of modern healthcare, yet traditional IV systems are prone to human errors, manual adjustments, and limited monitoring capabilities, which can jeopardize patient safety and hinder healthcare efficiency. This paper presents the development of the Smartdrip IV Infusion System, a sophisticated and automated solution that integrates advanced sensors, control algorithms, and intuitive user interfaces to address these limitations. The Smartdrip system minimizes human intervention, provides real-time monitoring, and automates fluid administration with enhanced precision and safety features. The system demonstrates high reliability through simulation and prototype testing in detecting anomalies, issuing alerts, and ensuring accurate infusion rates, thereby improving patient outcomes and reducing healthcare provider workload. The results validate the Smartdrip system as a transformative approach to optimizing IV therapy in diverse clinical settings. © 2024 IEEE.

关键词： Automated Infusion System Drip Rate Monitoring Healthcare Automation Intravenous Therapy Patient

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Normative Learning for Quality Assessment on Diffusion MRI 22

Unsupervised Normative Learning for Quality Assessment on Di...

引用

22nd IEEE International Symposium on Biomedical Imaging, ISBI 2025

作者： Yu, Jiahao Wang, Tenglong He, Yifei Pan, Yiang He, Jianzhong Wu, Ye School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China College of Information Engineering Zhejiang University of Technology Hangzhou China

ISBN: (纸本)9798331520526

Quality assessment of diffusion MRI (dMRI) aims to ensure that the presence of image artifacts does not affect the results of subsequent image analysis. Recently, a variety of deep learning methods have been proposed to enhance the quality and utility of these acquisitions. Still, they require a large amount of training data with artifact labeling. In this work, we developed an Unsupervised Quality Assessment tool (UNL-QA) for dMRI data that can handle a variety of artifacts such as ghosting, spike, noise and swap. Our method uses a vector quantized variational autoencoder to train typical patterns on a standard dMRI dataset, achieving 99.732% accuracy in detecting artifacts and standard data and 95% sensitivity in detecting artifacts on the training dataset. Performance across datasets is also excellent. It has also been demonstrated to be replicable on other datasets with different acquisition parameters. Without labeled artifact images necessary, UNL-QA allows fast and paves the way for efficient and effective artifact detection in large datasets. The code is available at https://***/yjh200319/***. © 2025 IEEE.

关键词： Damage detection

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：