检索结果-内蒙古大学图书馆

12th Mediterranean Conference on Embedded Computing, MECO 2023

作者： Catak, Ferhat Ozgur Cali, Umit Kuzlu, Murat Sarp, Salih University of Stavanger Dept. of Electrical Engineering and Computer Science Stavanger Norway Norwegian University of Science and Technology Department of Electric Power Engineering Trondheim Norway Old Dominion University Electrical Engineering Technology NorfolkVA United States Virginia Commonwealth University Electrical and Computer Engineering RichmondVA United States

ISBN: (纸本)9798350322910

With the rise of intelligent applications, such as self-driving cars and augmented reality, the security and reliability of wireless communication systems have become increasingly crucial. One of the most critical components of ensuring a high-quality experience is channel estimation, which is fundamental for efficient transmission and interference management in wire-less networks. However, using deep neural networks (DNNs) in channel estimation raises security and trust concerns due to their complexity and the need for more transparency in decision-making. This paper proposes a Monte Carlo Dropout (MCDO)-based approach for secure and trustworthy channel estimation in 5G networks. Our approach combines the advantages of traditional and deep learning techniques by incorporating conventional pilot-based channel estimation as a prior in the deep learning model. Additionally, we use MCDO to obtain uncertainty-aware predictions, enhancing the model's security and trustworthiness. Our experiments demonstrate that our proposed approach outper-forms traditional and deep learning-based approaches regarding security, trustworthiness, and performance in SG scenarios. © 2023 IEEE.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

Towards Developing a Transformer-Based Bangla Typing Error Correction Model: A Deep Learning-Based Approach

Towards Developing a Transformer-Based Bangla Typing Error C...

引用

Information and Communication Technology for Sustainable Development (ICICT4SD), International Conference on

作者： Shahidul Salim Taqiul Islam Rowzatul Zannat Naeem Mia Mohtasim Fuad Hasan Murad Dept. of Computer Science and Engineering Khulna University of Engineering & Technology Khulna Bangladesh Dept. of Computer Science and Engineering Chattogram University of Engineering & Technology Chattogram Bangladesh

Despite being the sixth most widely spoken language globally, the Bangla language faces a lack of effective and efficient automated methods to correct keyboard errors in written text. Such errors can result in communication misunderstandings and misinterpretations, particularly in the digital domain, where written text plays a crucial role in communication and is heavily relied upon. However, no keyboard error model for the Bangla language has been found due to the lack of available datasets. In this research work, we have proposed a probabilistic approach to generate a synthetic Bangla keyboard typing error correction (BKTEC) dataset. By utilizing the novel BKTEC dataset, we have fine-tuned a transformer-based model for Keyboard error correction. After training it to convergence, the proposed model has been tested on a test dataset, and we have found a word level accuracy of 88.56%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Texture Features Based Hybrid Multi Support Vector Machine Model for the diagnosis of Alzheimer's disease through Brain MRI Images

Texture Features Based Hybrid Multi Support Vector Machine M...

引用

2023 International Conference on Computational Intelligence, Networks and Security, ICCINS 2023

作者： Mallika, R.M. Rani, K Usha Hemalatha, K. Giri, M. Dept. of Cse Siddarth Institute of Engineering &technology Andhra Pradesh Puttur India Sri Padmavati Mahila Visva Vidyalayam Dept. of Computer Science Andhra Pradesh Tirupati India Madanapalle Institute of Technology & Science Dept. of CSE-AI Andhra Pradesh Madanapalle India Siddarth Institute of Engineering & Technology Department of Cse Andhra Pradesh Puttur India

ISBN: (纸本)9798350313796

Alzheimer's Disease (AD) is a complex neurodegenerative disorder that severely affects cognitive functions and poses significant challenges in early and accurate diagnosis. In recent years, machine learning techniques have shown remarkable potential in medical image analysis, particularly when applied to Brain MRI Images for AD diagnosis. This study proposes a novel approach utilizing a Texture Features based Hybrid Multi-Support Vector Machine Model to enhance the precision and early detection of Alzheimer's disease through Brain MRI Images. The proposed model combines the strength of texture features extracted from Brain MRI Images with a Hybrid Multi- Support Vector Machine framework. Texture features capture subtle patterns and structural variations within brain images, making them valuable in distinguishing different stages of Alzheimer's disease. By leveraging the optimization capabilities of the Hybrid Multi-Support Vector Machine, the model effectively handles high-dimensional data and complex patterns, contributing to improved diagnostic accuracy. To evaluate the model's performance, a comprehensive dataset of Brain MRI Images, encompassing AD patients, Mild Cognitive Impairment cases, and non-AD individuals, was employed. The experimental results demonstrate the effectiveness of the proposed Texture Features based Hybrid Multi-Support Vector Machine Model in accurately classifying Alzheimer's disease cases, outperforming traditional diagnostic methods. © 2023 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

ADNet: An Advanced Artificial Neural Network for Early Diagnosis of Alzheimer's Disease

ADNet: An Advanced Artificial Neural Network for Early Diagn...

引用

Sustainable Technologies for Industry 4.0 (STI), International Conference on

作者： Md. Arif Hasan Rifat Nusrat Jahan Tanha Taranum Muhammad Nazrul Islam Faiz Al Faisal Department of Computer Science and Engineering Green University of Bangladesh Dhaka Bangladesh Dept. of Computer Science and Engineering Military Institute of Science and Technology Dhaka Bangladesh

ISBN: (数字)9798331531973

ISBN: (纸本)9798331531980

Alzheimer's disease is a progressive neurological disorder, often referred to as dementia, that requires early and accurate diagnosis for effective treatment and management. It's a complex disease type that originates in the brain and only gets worse with timestamps, which is why it should be diagnosed early and accurately. Traditional diagnostic methods often fall short, posing challenges to healthcare professionals. This paper introduces a novel deep learning technique, ADNet, a specialized convolutional neural network (CNN) developed to classify Alzheimer's disease into four stages. For MRI image processing, we applied preprocessing techniques, including Green Fire Blue conversion and contrast enhancement using CLAHE, to improve image clarity. To address data imbalance, various data augmentation methods were implemented. During experimentation, pre-trained models like VGG16, MobileNet-V2, InceptionV3, and ResNet-50 achieved accuracies of 85%, 85%, 79 %, and 67 %, respectively, but fell short of our goals. Our ADNet model, however, attained an impressive 97.26% accuracy. This work offers significant contributions to Alzheimer's research and advances deep learning applications in medical imaging and neurological disease diagnosis.

关键词： Neurological diseases Deep learning Training Technological innovation Accuracy Magnetic resonance imaging Computational modeling Brain modeling Convolutional neural networks Alzheimer's disease

来源：评论

学校读者我要写书评

暂无评论

Contribution of Algorithm Visualizations to Students' Learning Skills: A Pedagogical Approach

Contribution of Algorithm Visualizations to Students' Learni...

引用

International Symposium on Digital Forensic and Security (ISDFS)

作者： Harem Ali Kheder Asaf Varol Dept. of Computer Engineering College of Engineering and Computer Sciences Lebanese French University Erbil Iraq Dept. of Engineering Management and Technology College of Engineering and Computer Science The University of Tennessee at Chattanooga Chattanooga TN US

ISBN: (数字)9798350330366

ISBN: (纸本)9798350330373

The Domain Algorithm visualizations are multimedia-based representations of algorithms that help students learn how they operate and behave. Algorithm visualizations provide numerous benefits to instructors and students, including an improved understanding of difficult algorithms in programming. By stripping down the complexities of algorithms and presenting them in simplified visual ways, students can obtain a more thorough knowledge of how they perform. Instructional approaches, such as guided instruction or discovery learning, are critical to the pedagogical effectiveness of algorithm representations. Guided instructions can considerably boost learners' overall performance in problem-solving activities that incorporate visualizations. At the same time, discovery education promotes exploration and experimentation, yet it may result in longer learning times and increased mental demands. A combination of these approaches is required for optimal effectiveness of instruction. In this paper, we investigate how the pedagogical effectiveness of algorithm visualizations should be determined by a variety of elements, including student characteristics, visual and cognitive processing ability, algorithm representation, and instructional context based on a literature review and interviews with field experts, and a bunch of recommendations is provided that has witnessed by authors during their long teaching periods.

关键词： Visualization Education Linguistics computer science education Complexity theory Security Problem-solving

来源：评论

学校读者我要写书评

暂无评论

Evaluating Machine Learning Methods for Bangla Text Emotion Analysis

Evaluating Machine Learning Methods for Bangla Text Emotion ...

引用

Innovation in Technology (APCIT), Asia Pacific Conference on

作者： Tahmina Akter Mst. Sharmin Akter Tanjim Mahmud Dilshad Islam Mohammad Shahadat Hossain Karl Andersson Dept. of Computer Science and Engineering Port City International University Chittagong Bangladesh Dept. of Computer Science and Engineering Rangamati Science and Technology University Rangamati Bangladesh Dept. of Physical and Mathematical Sciences Chattogram Veterinary and Animal Sciences University Bangladesh Dept. of Computer Science and Engineering University of Chittagong Chittagong Bangladesh Cybersecurity Laboratory Luleå University of Technology Luleå Sweden

ISBN: (数字)9798350361537

ISBN: (纸本)9798350361544

Text-based emotion identification goes beyond simple sentiment analysis by capturing emotions in a more nuanced way, akin to shades of gray rather than just positive or negative sentiments. This paper details our experiments with emotion analysis on Bangla text. We collected a corpus of user comments from various social media groups discussing socioeconomic and political topics to identify six emotions: sadness, disgust, surprise, fear, anger, and joy. We evaluated the performance of four widely used machine learning algorithms—RF, DT, k-NN, and SVM—alongside three popular deep learning algorithms—CNNs, LSTM, and Transformer Learning—using TF-IDF feature extraction and word embedding techniques. The results showed that among the machine learning algorithms, DT, RF, k-NN, and SVM achieved accuracy scores of 82%, 84%, 73%, and 83%, respectively. In contrast, the deep learning models CNN and LSTM both achieved higher performance with an accuracy of 85% and 86% respectively. These findings highlight the effectiveness of traditional ML and DL approaches in detecting emotions from Bangla social media texts, indicating significant potential for further advancements in this area.

关键词： Deep learning Support vector machines Technological innovation Sentiment analysis Machine learning algorithms Accuracy Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

Device Distribution Scheme of Random Access in Space-Air-Ground Integrated Network for Massive IoT 13

Device Distribution Scheme of Random Access in Space-Air-Gro...

引用

13th International Conference on Information and Communication Technology Convergence, ICTC 2022

作者： Oh, Joohyun Park, Joohan Youn, Jiseung Im, Jeong-Ju Cho, Sunghyun Major in Bio-Artificial Intelligence Hanyang University Dept. of Applied Artificial Intelligence Ansan Korea Republic of Hanyang University Dept. of Computer Science & Engineering Ansan Korea Republic of Major in Bio-Artificial Intelligence Hanyang University Dept. of Computer Science & Engineering Ansan Korea Republic of

ISBN: (数字)9781665499392

ISBN: (纸本)9781665499392

To support massive random access trials of devices with diverse QoS is a major challenge for massive machine-type communications. Space-air-ground integrated network(SAGIN) can be a promising solution for the congestion issues because the SAGIN can extend preamble pool. In this paper, we propose a device distribution scheme to maximize random access success rate while conserving delay of the packet in SAGIN. The proposed scheme consists of two phases. One is an algorithm in which the device selects the appropriate SAGIN infrastructure to maximize the random access success rate. The other is an algorithm in which devices that have failed previous random access select an infrastructure different from the previously selected infrastructure to resolve congestion. We compare the performance of the proposed algorithm with two benchmark algorithms in terms of random access success rate and delay. © 2022 IEEE.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Speeding Up EfficientNet: Selecting Update Blocks of Convolutional Neural Networks using Genetic Algorithm in Transfer Learning

arXiv

引用

arXiv 2023年

作者： Hasan, Md. Mehedi Ibrahim, Muhammad Ali, Md. Sawkat Dept. of Computer Science and Engineering East West University Dhaka Bangladesh Dept. of Computer Science and Engineering Univeristy of Dhaka Dhaka Bangladesh

The performance of convolutional neural networks (CNN) depends heavily on their architectures. Transfer learning performance of a CNN relies quite strongly on selection of its trainable layers. Selecting the most effective update layers for a certain target dataset often requires expert knowledge on CNN architecture which many practitioners do not posses. General users prefer to use an available architecture (e.g. GoogleNet, ResNet, EfficientNet etc.) that is developed by domain experts. With the ever-growing number of layers, it is increasingly becoming quite difficult and cumbersome to handpick the update layers. Therefore, in this paper we explore the application of genetic algorithm to mitigate this problem. The convolutional layers of popular pre-trained networks are often grouped into modules that constitute their building blocks. We devise a genetic algorithm to select blocks of layers for updating the parameters. By experimenting with EfficientNetB0 pre-trained on ImageNet and using Food-101, CIFAR-100 and MangoLeafBD as target datasets, we show that our algorithm yields similar or better results than the baseline in terms of accuracy, and requires lower training and evaluation time due to learning less number of parameters. We also devise a metric called block importance to measure each block’s efficacy as update block and analyze the importance of the blocks selected by our algorithm. © 2023, CC BY-NC-ND.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Few-Shot Learning with Feature Pairing and Mean Discrepancy

Few-Shot Learning with Feature Pairing and Mean Discrepancy

引用

Sustainable Computing and Smart systems (ICSCSS), International Conference on

作者： K K Singh K Hima Bindu dept. of Computer Science and Engineering NIT Andhra Pradesh and RGUKT Nuzvid India dept. of Computer Science and Engineering NIT AP India

Few-Shot Learning (FSL) is a sub-area of machine learning which mainly deals with data where there is a scarcity of training supervised samples. Few shot learning (FSL) more closely resembles the human brain in comparing new concepts to others based on prior experience rather than identifying it exactly. FSL aims to generalize the model across the tasks (in meta learning) opposed to the classical supervised learning which generalizes across the data points. In general the FSL models may suffer from underfitting because of scarcity of supervised samples and at the same time it causes overfitting as it is likely to memorize task specific features of the training set. This work aims to reduce such problems and is presented as a metric based model ”Few Shot Learning with Feature Pairing and Mean Discrepancy” (FL-FPMD). As the title suggests, feature pairing is one among various data augmentations. It is observed that flip augmentation is more suitable in the context of pairing the features within the given task. Memorizing task specific features is reduced by incorporating the discrepancy of mean distributions of the query and the support embedding in the loss function. The training and the evaluation is performed at the miniImageNet dataset and the results indicate that the proposed model outperforms the state-of-the-art models of similar complexity.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing Question-Answering with Knowledge Graph Retrieval and Generation using LLMs

Enhancing Question-Answering with Knowledge Graph Retrieval ...

引用

Artificial Intelligence and Quantum Computation-Based Sensor Application (ICAIQSA), International Conference on

作者： Supraj Gijre Rishi Agrawal Priyash Laddha Gunjan Keswani Dept. Computer Science and Engineering Shri Ramdeobaba College of Engineering and Management Nagpur India School of Computer Science and Engineering Ramdeobaba University Nagpur India

ISBN: (数字)9798331517953

ISBN: (纸本)9798331517960

This paper describes a novel technique to improving Large Language Models (LLMs) for document analysis that employs knowledge graphs and retrieval-augmented generation (RAG). We are working on constructing a chatbot system that can handle and analyze large documents from a variety of fields. Our approach addresses basic LLM issues including context maintenance and hallucination avoidance. The system combines document chunking, vector embedding, and similarity search with graph-based knowledge representation. Users can upload large papers and answer questions accurately. We show that integrating standard information retrieval approaches with graph-based storage and LLM capabilities improves context awareness and response accuracy across a wide range of document genres. This strategy is especially promising for complicated publications such as financial reports.

关键词： Text analysis Quantum computing Large language models Soft sensors Retrieval augmented generation Knowledge graphs Information retrieval Vectors Maintenance Standards

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：