检索结果-内蒙古大学图书馆

作者： Dai, Jin Zhang, Xinyu Shanghai Key Laboratory of Trustworthy Computing Shanghai China Shanghai China School of Software Engineering East China Normal University Shanghai China

We present an improved image caption generation model that incorporating multimodal attention mechanism. We use ResNet-101 to extract image features while incorporating channel attention mechanism and spatial attention mechanism. We use Faster R-CNN for object detection and use a multi-head attention structure consisting of spatial attention and self-attention. This allows our algorithm to improve the model's capability to learn and use the internal grammatical features of natural sentences. Moreover, we use GPU parallel computing to accelerate the entire model training. We apply our model and algorithm to early education scenarios: show and tell for kids. We compare our algorithm with the state-of-the-art deep learning algorithms. Our experimental results show that our model improves the captioning accuracy in terms of standard automatic evaluation metrics. © 2022 John Wiley & Sons, Ltd.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Proposed intelligence systems based on digital Forensics: Review paper

引用

Materials Today: Proceedings 2023年 80卷 2647-2651页

作者： Ismael Sahib, Hussein Qahatan AlSudani, Mustafa Hasan Ali, Mohammed Qassim Abbas, Haydar Moorthy, Kohbalan Mundher Adnan, Myasar Faculty of Computer Science and Information Technology UniversitiTun Hussein Onn Malaysia Computer Techniques Engineering Department Faculty of Information Technology Imam Ja'afar Al-sadiq University Najaf Iraq Faculty of Computer Systems & Software Engineering Universiti Malaysia Pahang Kuantan Malaysia Islamic University Najaf Iraq Faculty of Engineering School of Computing Universiti Teknologi Malaysia 81310 Johor Bahru Malaysia

The field of information security, in general, has seen shifts a traditional approach to an intelligence system. Moreover, an increasing of researchers to focus on propose intelligence systems and framework based on the forensic case studies because of the limitations of traditional methods such as analysis intensive data manually, intelligence visualization to make the evidence more understandable and intelligence system for store data. However, most of these intelligence systems are still facing different limitations. Furthermore, the primary goal of this work analysis popular intelligence system that was used based on forensic. Moreover, propose new algorithms and hybrid model which it's achieved good results in dif-ferent other fields to develop the forensic systems in the future. © 2021

关键词： Digital forensics

来源：评论

学校读者我要写书评

暂无评论

Adversarial Learning Under Hybrid Perturbations for Robust Acute Lymphoblastic Leukemia Classification 39

Adversarial Learning Under Hybrid Perturbations for Robust A...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Chen, Jie Liu, Xinyuan Liu, Xintong Li, Jianqiang College of Computer Science and Software Engineering Shenzhen University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China

ISBN: (纸本)157735897X

Acute lymphoblastic leukemia is a childhood cancer prevalent worldwide, which can prove fatal within weeks or months. However, current diagnosis models based on machine learning and deep learning methods fail to consider device noise (pixel-level perturbations) and rotation/translation (spatial-transformed perturbations), which can undermine the model's robustness. Adversarial training is a potential solution to this issue. This paper presents a hybrid perturbation adversarial training (HPAT) strategy that leverages two types of adversarial samples: pixel-level adversarial samples and spatial adversarial samples. This work generates these hybrid adversarial samples through Projected Gradient Descent (PGD) in couple with spatial transformation based on the Bayesian optimization (STBO) algorithm, respectively. This work introduced the Mixed Batch Normalization (MixBN) module to handle both adversarial samples and clean samples, alleviating the problem of clean accuracy degradation due to adversarial training. The proposed hybrid adversarial training strategy is tested on the public acute lymphoblastic leukemia dataset and found that it outperformed existing acute lymphoblastic cell classification models. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Deepfakes – Reality Under Threat?

Deepfakes – Reality Under Threat?

引用

IEEE Annual computing and Communication Workshop and Conference (CCWC)

作者： Md Shohel Rana Md Solaiman Charan Gudla Md Fahimuzzman Sohan Computing and Software Engineering Florida Gulf Coast University Fort Myers United States Software Engineering Daffodil International University Dhaka Bangladesh Computer Science and Engineering Mississippi State University Starkville United States

In recent years, advances in technology, specifically in machine learning, artificial intelligence, and deep learning, have made it possible to create highly convincing fake videos known as Deepfakes. These videos are produced by training computer models on extensive datasets of faces and then seamlessly blending one person’s facial expressions onto another’s, resulting in videos that are nearly indistinguishable from reality. The widespread use of Deepfakes presents several concerns, including the creation of political distress, the occurrence of fake terrorism events, the undermining of trust in digital media, etc. Therefore, there is an urgent need to continually advance Deepfake detection and prevention methodologies to safeguard against their malevolent use and maintain the integrity of digital content. This paper conducts a meticulous analysis of the current landscape of Deepfake research, surveys the most effective detection solutions, and introduces a real-time Deepfake detection and prevention model within a rigorous testing framework. This model integrates innovative Blockchain and Steganalysis technologies to provide a robust solution to combat the explosion of Deepfakes. Our holistic framework offers a systematic and statistically rigorous approach to distinguishing genuine content from its manipulated counterparts, Deepfakes. By employing the principles of hypothesis testing, and a robust test statistic, our research equips us with the analytical tools necessary to make well-informed and precise classifications, significantly contributing to the ongoing battle against Deepfakes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Workload Prediction of Virtual Machines Using Integrated Deep Learning Approaches Over Cloud Data Centers 1st

Workload Prediction of Virtual Machines Using Integrated Dee...

引用

1st International Conference on Human-Centric Smart computing, ICHCSC 2022

作者： Leka, Habte Lejebo Fengli, Zhang Kenea, Ayantu Tesfaye Sharma, Durga P. Tegene, Abebe Tamrat School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China Department of Computer Science and Engineering School of Electrical Engineering and Computing Adama Science and Technology University Adama1888 Ethiopia AMUIT MOEFDRE Under UNDP & MAISM RTU Kota Kota India

ISBN: (纸本)9789811954023

Exponential growth in the use of cloud computing services makes it difficult to forecast loads of virtual machines (VMs). Accurate virtual machine (VM) workload forecasting is the most critical task in appropriately managing cloud resources such as memory and central processing units while minimizing energy usage. To address this problem, an integrated deep learning model is proposed in this research paper. The model employs two popular neural networks: a bidirectional Long Short-Term Memory Network (BiLSTM) with a convolutional neural networks (CNN). The CNN component pulls high-level attributes from all VM workload data, whereas the BiLSTM component forecasts future VM workload. The experimental results reveal that the suggested model outperforms commonly used workload prediction methods in terms of forecasting accuracy of VMs workloads in cloud computing environments. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Designing Uncorrelated Address Constrain for DNA Storage by DMVO Algorithm

引用

IEEE/ACM Transactions on Computational Biology and Bioinformatics 2022年第2期19卷 866-877页

作者： Cao, Ben Ii, Xue Zhang, Xiaokang Wang, Bin Zhang, Qiang Wei, Xiaopeng Key Laboratory of Advanced Design and Intelligent Computing Ministry of Education Ministry of Education School of Software Engineering Dalian University Dalian116622 China School of Computer Science and Technology Dalian University of Technology Dalian116024 China

At present, huge amounts of data are being produced every second, a situation that will gradually overwhelm current storage technology. DNA is a storage medium that features high storage density and long-term stability and is now considered to be a feasible storage solution. Errors are easily made during the sequencing and synthesis of DNA, however. In order to reduce the error rate, novel uncorrelated address constrain are reported, and a Damping Multi-Verse Optimizer (DMVO)algorithm is proposed to construct a set of DNA coding, which is used as the non-payload. The DMVO algorithm exchanges objects through black/white holes in order to achieve a stable state and adds damping factors as disturbances. Compared with previous work, the coding set obtained by the DMVO algorithm is larger in size and of higher quality. The results of this study reveal that the size of the DNA storage coding set obtained by the DMVO algorithm increased by 4-16 percent, and the variance of the melting temperature decreased by 3-18 percent. © 2004-2012 IEEE.

关键词： Damping

来源：评论

学校读者我要写书评

暂无评论

Hyacinth macaw: a project-based learning program to develop talents in software engineering for Artificial Intelligence 23

Hyacinth macaw: a project-based learning program to develop ...

引用

37th Brazilian Symposium on software engineering, SBES 2023, held in conjunction with the Brazilian Conference on software: Theory and Practice, CBSoft 2023

作者： Fontão, Awdren Matsubara, Edson Mongelli, Henrique Medeiros, Marcio Lourenço, Carlos Martins, Henrique Cortez, Igor Borges, Maria Faculty of Computing / Software Engineering Lab Federal University of Mato Grosso Do Sul Brazil PDtec SP Brazil B3 SP Brazil Blk SP Brazil Neoway SP Brazil

ISBN: (纸本)9798400707872

software engineering for Artificial Intelligence (SE4A) uses SE principles to design and maintain AI systems, requiring analytical thinking for software complexity, while AI demands mathematical knowledge and algorithm adjustment. The IEEE Curriculum Guidelines for Undergraduate Degree Programs in software engineering states that extracurricular elements impact students' preparation. This study focuses on the first module of a project-based learning talent development program involving undergraduate students, two expert professors (in AI and SE), and mentors from sponsoring companies. An exploratory case study with 39 students from four courses was conducted, challenging them to deliver an MVP in machine learning within 1.5 months. Results showed high agreement (87.5%) in applying learned skills to future projects, recognizing SE's benefits (96.9%) in AI, and acknowledging the connection between SE and AI (78.1%). Participants applied relevant knowledge in ML performance, data analysis, and software architecture for AI. We share strategies used by students to enhance developer experience. © 2023 ACM.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

From Data Literacy to Co-design Environmental Monitoring Innovations and Civic Action

From Data Literacy to Co-design Environmental Monitoring Inn...

引用

International Conference on Wireless Communications, Networking and Applications, WCNA 2021

作者： Happonen, Ari Wolff, Annika Palacin, Victoria Software Engineering School of Engineering Science LUT University Lappeenranta53850 Finland Social Computing Research Group Faculty of Social Sciences University of Helsinki Helsinki00014 Finland

ISBN: (纸本)9789811924552

SENSEI is an environmental monitoring initiative run by Lappeenranta University of Technology (LUT University) and the municipality of Lappeenranta in south-east Finland. The aim was to collaboratively innovate and co-design, develop and deploy civic technologies with local civics to monitor positive and negative issues. These are planned to improve local’s participation to social governance issues in hand. These issues can be e.g. waste related matters like illegal dumping of waste, small vandalism into city properties, alien plant species, but on the other hand nice places to visits too. This publication presents initiatives data literacy facet overview, which is aimed at creating equitable access to information from open data, which in turn is hoped for to increase participants motivation and entrepreneurship like attitude to work with the municipals and the system. This is done by curating environmental datasets to allow participatory sensemaking via exploration, games and reflection, allowing citizens to combine their collective knowledge about the town with the often-complex data. The ultimate aim of this data literacy process is to enhance collective civic actions for the good of the environment, to reduce the resource burden in the municipality level and help citizens to be part of sustainability and environmental monitoring innovation activities. For further research, we suggest follow up studies to consider on similar activities e.g. in specific age groups and to do comparisons on working with different stage holders to pin point most appropriate methods for any specific focus group towards collaborative innovation and co-design of civic technologies deployment. © 2022, The Author(s).

关键词： Open Data

来源：评论

学校读者我要写书评

暂无评论

Sinhala Sign Language Detection Approach for Deaf People Using Human Pose Estimation

Sinhala Sign Language Detection Approach for Deaf People Usi...

引用

International Research Conference on Smart computing and Systems engineering (SCSE)

作者： S. J. M. J. Nadeesha W. V. S. K. Wasalthilaka Department of Computing and Information Systems Faculty of Computing Sabaragamuwa University of Sri Lanka Belihuloya Sri Lanka Department of Software Engineering Faculty of Computing Sabaragamuwa University of Sri Lanka Belihuloya Sri Lanka

ISBN: (数字)9798350375688

ISBN: (纸本)9798350375695

People with disabilities have physical, sensory, cognitive, or behavioral health problems that limit their participation in activities and interaction with the environment. Deafness is one of the most common types of disability in society. Since verbal communication is impossible for the deaf, sign language is one of the most often used forms of communication among those disable people. However, a third party who is fluent in sign language is necessary for efficient communication. This research study proposed a sign language detector for Sinhala using Logistic Regression, Ridge Classifier, Random Forest, and Gradient Boosting algorithms with human pose estimation. The method used upper-body stance landmarks with 100 human pose videos and 40 widely used signals to extract features for body language classification. With an overall 86.75% success rate, this study is a major step toward improving deaf people's communication in Sri Lanka.

关键词： Training Sign language Logistic regression Pose estimation Detectors Feature extraction Systems engineering and theory

来源：评论

学校读者我要写书评

暂无评论

TreeBERT: Advanced Representation Learning for Relation Extraction

TreeBERT: Advanced Representation Learning for Relation Extr...

引用

2023 IEEE International Conference on Big Data, BigData 2023

作者： Hebbar, Shashank Xie, Ying Zhong, Jiang Risk Modeling and Analytics Credigy Solutions Inc Norcross United States Kennesaw State University College of Computing and Software Engineering Kennesaw United States Chongqing University College of Computer Science Chongqing China

ISBN: (纸本)9798350324457

As the COVID-19 pandemic has subsided, it remains crucial to analyze the vast amount of research produced during this period to advance our understanding of effective medical treatments and therapies. The wealth of biomedical literature generated during the pandemic presents a challenge for researchers who need to quickly assimilate new information. To tackle this issue, we introduce Tree-BERT, a novel representation learning language model designed for the automatic extraction of biomedical relationships from medical publications. Unlike traditional BERT models and their variants, which solely focus on the contextual relationship between words in input sentences, Tree-BERT also incorporates sentence structure information, enhancing word embeddings. Our experiments reveal that Tree-BERT outperforms alternative methods in biomedical relationship extraction. We will also demonstrate the model's ability to identify new relationships from Covid-19 publications, showcasing its potential to contribute to the ongoing analysis of pandemic research and its broader applications in the field of bio-medicine. © 2023 IEEE.

关键词： BERT Biomedical Relation Extraction Graph Convolutional Network Representation Learning TreeBERT

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：