检索结果-内蒙古大学图书馆

14th International Conference on computing Communication and Networking Technologies, ICCCNT 2023

作者： Vinayak, Vineet Jyotsna, C. Department of Computer Science and Engineering Amrita School of Computing Bengaluru India

ISBN: (纸本)9798350335095

The goal of Text classification is to categorize a document into predefined categories. Various supervised and unsupervised classifiers can be used to achieve this. In the research work proposed, SOTA (State of the Art) deep learning models and embedding techniques are used to classify consumers' complaints, which is in form of text, into 6 classes. Words and documents are represented as numeric vectors through embedding, enabling vector representations for related words. These representations are easily ingested by NLP algorithms. The classes represent departments where complaints are routed. Deep learning models like LSTM, Bi-LSTM, GRU and 1D CNN are used, along with word embedding techniques like Word2Vvec, Fasttext, Bert and Distilbert to represent text. The experimental results indicate that DistilBert and CNN achieved a 93% F-score. © 2023 IEEE.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

BiT5: A Bidirectional NLP Approach for Advanced Vulnerability Detection in Codebase 5

BiT5: A Bidirectional NLP Approach for Advanced Vulnerabilit...

引用

5th International Conference on Innovative Data Communication Technologies and Application, ICIDCA 2024

作者： Prabith, G.S. Rohit Narayanan, M. Arya, A. Aneesh Nadh, R. Binu, P.K. Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Amritapuri India Department of Computer Science and Applications Amrita School of Computing Amrita Vishwa Vidyapeetham Amritapuri India

In this research paper, a detailed investigation presents the utilization of the BiT5 Bidirectional NLP model for detecting vulnerabilities within codebases. The study addresses the pressing need for techniques enhancing software security by effectively identifying vulnerabilities. Methodologically, the paper introduces BiT5, specifically designed for code analysis and vulnerability detection, encompassing dataset collection, preprocessing steps, and model fine-tuning. The key findings underscore BiT5's efficacy in pinpointing vulnerabilities within code snippets, notably reducing both false positives and false negatives. This research contributes by offering a methodology for leveraging BiT5 in vulnerability detection, thus significantly bolstering software security and mitigating risks associated with code vulnerabilities. © 2024 The Authors. Published by Elsevier B.V.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Segmenting the Brain Tumor using Deep Learning Techniques 3

Segmenting the Brain Tumor using Deep Learning Techniques

引用

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Pranitha, Ch. Lakshmi Sumathi, R. Harshitha, V. Kavya, N. Soni, A. Kalasalingam Academy of Research and Education Department of Computer Science and Engineering Tamil Nadu India Kalasalingam Academy of Research and Education School of Computing Department of Computer Science and Engineering Tamil Nadu India

ISBN: (纸本)9798331509675

This study presents a U-Net architecture with EfficientNet-B7 as the feature extraction backbone is used to propose a robust brain tumor segmentation method. Using cutting-edge deep learning techniques, the suggested model precisely segments MRI images with an emphasis on identifying and defining tumor anomalies. A thorough preprocessing pipeline that handles missing values, augments data with albumentations, and normalizes data guarantees the consistency and resilience of the incoming data. The segmentation procedure uses a hybrid loss function that combines Soft Dice Loss and Binary Cross-Entropy (BCE), improving the model's ability to strike a compromise between segmentation consistency and pixel-level precision. To ensure efficiency and generalizability, the model training is optimized using the Adam optimizer in conjunction with a learning rate scheduler. Additionally, an early halting mechanism is incorporated to minimize overfitting. Using a standardized dataset, performance evaluation is carried out with 70%, 15%, and 15% splits for training, validation, and testing, respectively. To evaluate the quality of segmentation, important metrics are employed, such as the Dice Coefficient and the Intersection over Union (IoU). The accuracy of the model in identifying tumor boundaries is illustrated by visualizations of the anticipated segmentation results. The potential of combining U-Net with cutting-edge feature extractors such as EfficientNet-B7 to produce dependable, effective, and clinically significant segmentation results in medical imaging is demonstrated by this approach. © 2025 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Ortho Vision: Autoencoder-CNN Fusion Approach for Fracture Detection 7

Ortho Vision: Autoencoder-CNN Fusion Approach for Fracture D...

引用

7th International Conference on Electronics, Communication and Aerospace Technology, ICECA 2023

作者： Abhiram, S. Amal Krishna, K. Vaishnav, G.M. Anjali, T. Abhishek, S. Amrita School of Computing Department of Computer Science and Engineering Amritapuri India

ISBN: (纸本)9798350340600

We propose an advanced framework for automated fracture detection in medical X-ray images., harnessing the power of hybrid deep learning methodologies. Through the fusion of autoencoders' adeptness in feature extraction and a convolutional neural networks (CNN) efficiency in recognizing the complex patterns., the proposed model achieves remarkable accuracy in identifying fractures. By integrating data augmentation and preprocessing strategies., the system demonstrates enhanced robustness in handling variations in image quality and presentation. This innovation holds substantial promise in revolutionizing the field of medical diagnostics by offering a reliable and efficient tool for clinicians to expedite fracture diagnosis., facilitate treatment decisions., and ultimately improve patient care. The amalgamation of diverse techniques not only establishes a robust fracture detection mechanism but also showcases the potential of interdisciplinary approaches. The proposed approach demonstrated excellent accuracy in fracture detection with a 92% percentage., which can greatly improve medical diagnosis and treatment planning. © 2023 IEEE.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

Machine Learning in Lithium-Ion Battery: Applications, Challenges, and Future Trends

引用

SN computer science 2024年第6期5卷 717页

作者： Valizadeh, Alireza Amirhosseini, Mohammad Hossein NEX Power Ltd Department of Computer Science and Digital Technologies School of Architecture Computing and Engineering University of East London

Machine Learning has garnered significant attention in lithium-ion battery research for its potential to revolutionize various aspects of the field. This paper explores the practical applications, challenges, and emerging trends of employing Machine Learning in lithium-ion battery research. Delves into specific Machine Learning techniques and their relevance, offering insights into their transformative potential. The applications of Machine Learning in lithium-ion-battery design, manufacturing, service, and end-of-life are discussed. The challenges including data availability, data preprocessing and cleaning challenges, limited sample size, computational complexity, model generalization, black-box nature of Machine Learning models, scalability of the algorithms for large datasets, data bias, and interdisciplinary nature and their mitigations are also discussed. Accordingly, by discussing the future trends, it provides valuable insights for researchers in this field. For example, a future trend is to address the challenge of small datasets by techniques such as Transfer Learning and N-shot Learning. This paper not only contributes to our understanding of Machine Learning applications but also empowers professionals in this field to harness its capabilities effectively. © The Author(s) 2024.

关键词： Artificial intelligence Data-driven approach Lithium-ion battery Machine learning

来源：评论

学校读者我要写书评

暂无评论

An Ensemble based Convolutional Neural Network modelling for Classifying Medulloblastoma subtype 5

An Ensemble based Convolutional Neural Network modelling for...

引用

5th IEEE International Conference for Emerging Technology, INCET 2024

作者： Sathwik, T. Reddy, Tadi Sai Vishruth Charitha, Talluri Singh, Rimjhim Padam Kanchan, Sneha Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Bengaluru India Department of Internet Engineering and Computer Science Universiti Tunku Abdul Rahman Malaysia

ISBN: (纸本)9798350361155

Medulloblastoma a heterogeneous pediatric brain tumor, requires specific subtype identification in order to determine the most effective treatment strategies. To evaluate the accuracy of classification, this study examines state-of-art leading Convolutional Neural Network (CNN), models like ResNet-50, MobileNet, Xception model, ensembles, etc in differentiating between medulloblastoma subtypes using histopathological images. The models undergo rigorous training and fine-tuning using a comprehensive benchmark dataset encompassing a variety of medulloblastoma subtypes. We also employ advanced image pre-processing techniques to optimize feature extraction. Through comparative analyses based on the accuracy, sensitivity, and specificity and precision of each model to uncover their unique strengths and weaknesses, the work identifies the top performing model for medulloblastoma subtypes classification. Then considering unique strengths of these models, we propose an ensemble methodology that combines top performing models for the underlying problem. Through experimentation, our ensemble approach proves to have the highest accuracy in medulloblastoma subtype classification. The combination of ResNet-50's depth, MobileNet's efficiency, and Xception's feature extraction abilities contribute to the ensemble's outstanding performance with 97% of acuuracy and F-score. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Block chain-based decentralized public auditing for cloud storage with improved EIGAMAL encryption model

引用

International Journal of Information Technology (Singapore) 2024年第2期16卷 697-711页

作者： Kumar, R. Prasanna Bandanadam, Showri Rayalu Department of Computer Science and Engineering (AIE) Amrita School of Computing Amrita Vishwa Vidyapeetham Tamil Nadu Chennai India

With the growing relevance of data integrity, public auditing mechanisms for cloud storage models are intensively researched. In public auditing systems, a third person is used to ensure the outsourced data’s integrity on the user’s behalf. A lot of public verification models regarding blockchain are developed to combat cruel trusted Third Public Auditors (TPAs). Conventional models for auditing are based on TPA and are prone to an auditor enticing who colludes with the hostile block-chain to provide prejudiced results for auditing. This research work plans to propose a blockchain-based public auditing scheme to verify data integrity in cloud storage. The first people upload the data to the cloud and retrieve the data when they need it. The data must be encrypted and private among the second people and DO, since both DO and the second person do not believe each other. Therefore, the DO encrypts his/her data using the Improved EI-GAMAL (IEI-GAMAL) encryption algorithm and uploads the encrypted data to the Cloud Service Provider (CSP) via the blocks of the blockchain. For preserving the privacy of the data, the key for data encryption as well as decryption is generated using the new hybrid optimization model. The DO selects one TPA to audit the data on behalf of the DO. Generally, a TPA concentrates the data of numerous DOs and offers the audit for DOs. The DO requests a public audition to generate proof of the encrypted block stored in the blockchain when he requires verifying the data present in cloud storage. In order to provide strong updates and lessen the computation expenses, the new hybrid optimization algorithm is used. The new hybrid algorithm is the theoretical unify of the Adaptive Position-based Arithmetic with the Sunflower Optimization Algorithm (APA-SFOA). Several investigations are executed to prove the supremacy of the developed model. © The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of computer Applications and Management 2

关键词： Adaptive position-based arithmetic with sunflower optimization algorithm Blockchain technology Improved EI-GAMAL Public auditing for cloud storage

来源：评论

学校读者我要写书评

暂无评论

Optimizing University Time Table Schedules for Smart Space Utilization 5

Optimizing University Time Table Schedules for Smart Space U...

引用

5th IEEE Global Conference for Advancement in Technology, GCAT 2024

作者： Dasari, Chetan Aditya Kumar, Solasa Manikanta Arpitalaxmi, Arpitalaxmi Thushara, M.G. Amrita Vishwa Vidyapeetham Amrita School of Computing Department of Computer Science and Engineering Amritapuri India Amrita Vishwa Vidyapeetham Amrita School of Computing Department of Computer Science and Applications Amritapuri India

ISBN: (纸本)9798350376685

Manual university course timetable scheduling is a time-consuming and error-prone process, often resulting in conflicts and inefficiencies. An optimized timetable is crucial for smooth academic operations, ensuring efficient use of resources and fostering a positive learning environment for students and faculty. We address the challenge of university course scheduling by proposing an optimized time-table scheduling algorithm. This paper will identify suitable approaches to automate timetable generation, considering factors like lecturer availability, student groups, room capacity, and course requirements. Keeping track of all the different factors that affect things like course requirements, teacher availability, room capacity, and fair task distribution is hard. By working on slot-based scheduling methods, we hope to find ways to make the schedule work best for both students and teachers, which will make everyone happier and help them learn more. © 2024 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

An Iterative PRISMA Review of GAN Models for Image Processing, Medical Diagnosis, and Network Security

引用

computers, Materials & Continua 2025年第2期82卷 1757-1810页

作者： Uddagiri Sirisha Chanumolu Kiran Kumar Sujatha Canavoy Narahari Parvathaneni Naga Srinivasu Department of Computer Science and Engineering Prasad V Potluri Siddhartha Institute of TechnologyVijayawada520007India Department of Computer Science and Engineering Koneru Lakshmaiah Education FoundationVaddeswaram522302India Department of Electronics and Communication Engineering Sreenidhi Institute of Science and TechnologyHyderabad501301India Amrita School of Computing Amrita Vishwa VidyapeethamAmaravati522503India Department of Teleinformatics Engineering Federal University of CearáFortaleza60455-970Brazil Inti International University Nilai71800Malaysia

The growing spectrum of Generative Adversarial Network (GAN) applications in medical imaging, cyber security, data augmentation, and the field of remote sensing tasks necessitate a sharp spike in the criticality of review of Generative Adversarial Networks. Earlier reviews that targeted reviewing certain architecture of the GAN or emphasizing a specific application-oriented area have done so in a narrow spirit and lacked the systematic comparative analysis of the models’ performance metrics. Numerous reviews do not apply standardized frameworks, showing gaps in the efficiency evaluation of GANs, training stability, and suitability for specific tasks. In this work, a systemic review of GAN models using the PRISMA framework is developed in detail to fill the gap by structurally evaluating GAN architectures. A wide variety of GAN models have been discussed in this review, starting from the basic Conditional GAN, Wasserstein GAN, and Deep Convolutional GAN, and have gone down to many specialized models, such as EVAGAN, FCGAN, and SIF-GAN, for different applications across various domains like fault diagnosis, network security, medical imaging, and image segmentation. The PRISMA methodology systematically filters relevant studies by inclusion and exclusion criteria to ensure transparency and replicability in the review process. Hence, all models are assessed relative to specific performance metrics such as accuracy, stability, and computational efficiency. There are multiple benefits to using the PRISMA approach in this setup. Not only does this help in finding optimal models suitable for various applications, but it also provides an explicit framework for comparing GAN performance. In addition to this, diverse types of GAN are included to ensure a comprehensive view of the state-of-the-art techniques. This work is essential not only in terms of its result but also because it guides the direction of future research by pinpointing which types of applications require some

关键词： GAN CGAN WGAN DCGAN image analysis

来源：评论

学校读者我要写书评

暂无评论

A survey for light field super-resolution

引用

High-Confidence computing 2024年第1期4卷 118-129页

作者： Mingyuan Zhao Hao Sheng Da Yang Sizhe Wang Ruixuan Cong Zhenglong Cui Rongshan Chen Tun Wang Shuai Wang Yang Huang Jiahao Shen State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and EngineeringBeihang UniversityBeijing 100191China Key Laboratory of Data Science and Intelligent Computing International Innovation InstituteBeihang UniversityHangzhou 311115China State Key Laboratory of Software Development Environment School of Computer Science and EngineeringBeihang UniversityBeijing 100191China

Compared to 2D imaging data,the 4D light field(LF)data retains richer scene’s structure information,which can significantly improve the computer’s perception capability,including depth estimation,semantic segmentation,and LF ***,there is a contradiction between spatial and angular resolution during the LF image acquisition *** overcome the above problem,researchers have gradually focused on the light field super-resolution(LFSR).In the traditional solutions,researchers achieved the LFSR based on various optimization frameworks,such as Bayesian and Gaussian *** learning-based methods are more popular than conventional methods because they have better performance and more robust generalization *** this paper,the present approach can mainly divided into conventional methods and deep learning-based *** discuss these two branches in light field spatial super-resolution(LFSSR),light field angular super-resolution(LFASR),and light field spatial and angular super-resolution(LFSASR),***,this paper also introduces the primary public datasets and analyzes the performance of the prevalent approaches on these ***,we discuss the potential innovations of the LFSR to propose the progress of our research field.

关键词： Light field super-resolution Convolutional neural network Transformer Sub-aperture image Epipolar-plane image

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：