检索结果-内蒙古大学图书馆

International Conference on Computational Intelligence and Applications (ICCIA)

作者： Abhibhu Tachaapornchai Somkiat Kosolsombat Chiabwoot Ratanavilisagul Data Science and Innovation College of Interdisciplinary Studies Thammasat University Pathum Thani Thailand Department of Computer and Information Science Faculty of Applied Science King Mongkut's University of Technology North Bangkok Thailand

ISBN: (数字)9798350352214

ISBN: (纸本)9798350352221

To build an intelligent intrusion detection system, it is essential to have a suitable and high-quality dataset with a sufficiently large quantity to simulate real-world scenarios. The NSL-KDD dataset is an improved version derived from the previous KDD 99 dataset. In this article, an analysis of the NSL-KDD dataset was conducted to study the efficiency of classification machine learning in detecting abnormalities in network data transmission patterns. The study achieved the highest accuracy by XGBoost method using 14 features important and grid search for hyperparameter tuning and revealing many interesting insights. This research has significant implications and can be further applied in the field of cybersecurity by leveraging machine learning for network intrusion detection systems.

关键词： Accuracy Computational modeling Machine learning NSL-KDD Feature extraction Computer networks Decision trees data communication Computer security Tuning

来源：评论

学校读者我要写书评

暂无评论

PDF Malware Detection Using Machine Learning

PDF Malware Detection Using Machine Learning

引用

International Conference on Computational Intelligence and Applications (ICCIA)

作者： Naragorn Tantikulvichit Somkiat Kosolsombat Chiabwoot Ratanavilisagul Data Science and Innovation College of Interdisciplinary Studies Thammasat University Pathum Thani Thailand Department of Computer and Information Science Faculty of Applied Science King Mongkut's University of Technology North Bangkok Thailand

ISBN: (数字)9798350352214

ISBN: (纸本)9798350352221

In today's world, cyber-attacks are on the rise, and PDF files are commonly used as a means of attack. One common type of attack through PDF files is the covert embedding of dangerous malware and tricking users into clicking on malicious links. The victims may have their important data stolen and suffer various forms of damage. Detecting malware embedded in PDF files can help mitigate the harm to users. This project aims to study methods for detecting malware embedded in PDF files using machine learning techniques, with the best-performing model being Gradient Boosting, achieving an accuracy of 99% in 5-fold cross-validation and 97% using 10 features. Using Feature Selection Technique Voting from 3 Models consist of Logistic Regression, Random Forest, Lasso Regression, Each model will choose matching features to use for machine learning analysis. This project approach reduce feature for machine learning to detect PDFMalware.

关键词： Training Analytical models Logistic regression Machine learning Feature extraction Boosting Malware data models Random forests Cyberattack

来源：评论

学校读者我要写书评

暂无评论

Review of Applications in Brain-Computer Interfaces Using the EMOTIV Insight Headset

Review of Applications in Brain-Computer Interfaces Using th...

引用

technology & Engineering Management Conference - Asia Pacific (TEMSCON-ASPAC), IEEE

作者： Yamin Thwe Nipat Jongsawat Julaluk Watthananon Padma Nyoman Crisnapati I Made Agus Wirawan Yuri Pamungkas Computer Science Department Faculty of Science and Technology RMUTT PathumThani Thailand Data and Information Science Faculty of Science and Technology RMUTT PathumThani Thailand Mechatronics Engineering Department Faculty of Technical Education RMUTT PathumThani Thailand Education of Informatic Engineering Department Universitas Pendidikan Ganesha Bali Indonesia Medical Technology Institut Teknologi Sepuluh Nopember Surabaya Indonesia

ISBN: (数字)9798331515881

ISBN: (纸本)9798331515898

This review explores advancements in Brain-Computer Interfaces (BCIs) using the EMOTIV Insight headset, a non-invasive electroencephalography (EEG) device. The study addresses the rising interest in affordable EEG devices for applications like robotic control, emotional and cognitive analysis, assistive technologies, and security in learning models. By synthesizing findings from multiple studies, the review emphasizes the EMOTIV Insight's strengths, including ease of use and wireless connectivity, alongside its limitations in signal fidelity and artifact management due to its five-channel configuration. Studies from 2016 to 2024 showcase the device's role in fields such as neurofeedback, mental health monitoring, and assistive technology development. Key findings highlight BCIs' success in controlling robotic systems, enhancing emotional and cognitive analyses, and improving accessibility for individuals with disabilities. However, challenges such as signal noise, user variability, and the need for advanced machine learning algorithms are also discussed. The review concludes that while EMOTIV Insight is a valuable tool for BCI research, addressing its challenges is essential to unlock its full potential. Continued advancements could significantly enhance the quality of life for individuals with disabilities and provide deeper insights into human cognition and emotion.

关键词： Headphones Wireless communication Reviews Mental health Assistive technologies Control systems Electroencephalography Brain-computer interfaces Robots System analysis and design

来源：评论

学校读者我要写书评

暂无评论

Unlocking the Future: A Decade of Insights into Short-Term Stock Market Trends with Machine Learning and the S&P 500 Index

Unlocking the Future: A Decade of Insights into Short-Term S...

引用

Innovative Research in Applied science, Engineering and technology (IRASET),Conference on

作者： Salma Tajbel Zouhair Elamrani Abou Elassad Abdelghafour Atlas Modélisation des systèmes complexes (MSC) Ecole nationale des sciences appliquées Marrakech Morocco Data Science Lab Faculty of Science and Information Technology Birulia Bangladesh

ISBN: (数字)9798350309508

ISBN: (纸本)9798350309515

This research paper presents a comprehensive exploration of short-term stock market trend prediction using state-of-the-art machine learning techniques, anchored in a decade-long analysis of the S&P 500 Index. Leveraging a rich dataset spanning ten years, our study investigates the performance of various machine learning models, including Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNN), Support Vector Machines (SVM), Random Forest, and Gated Recurrent Unit (GRU) models. These models were evaluated based on their ability to harness financial indicators and trading volume data, providing a deep dive into the complex factors influencing stock market movements. The GRU model, in particular, demonstrated exceptional performance, underscoring its potential as a superior predictive tool due to its efficient handling of temporal dependencies. This comparative analysis not only highlights the importance of selecting the appropriate model to match specific data characteristics but also contributes significantly to the broader understanding of how machine learning can enhance the prediction of stock market dynamics.

关键词： Support vector machines Analytical models Predictive models Market research data models Convolutional neural networks Indexes

来源：评论

学校读者我要写书评

暂无评论

Evolutionary Approach to Security Games with Signaling 31

Evolutionary Approach to Security Games with Signaling

引用

31st International Joint Conference on Artificial Intelligence, IJCAI 2022

作者： Żychowski, Adam Mańdziuk, Jacek Bondi, Elizabeth Venugopal, Aravind Tambe, Milind Ravindran, Balaraman Faculty of Mathematics and Information Science Warsaw University of Technology Poland Center for Research on Computation and Society Harvard University United States Robert Bosch Centre for Data Science and AI IIT Madras India Department of Computer Science and Engineering IIT Madras India

ISBN: (纸本)9781956792003

Green Security Games have become a popular way to model scenarios involving the protection of natural resources, such as wildlife. Sensors (e.g. drones equipped with cameras) have also begun to play a role in these scenarios by providing real-time information. Incorporating both human and sensor defender resources strategically is the subject of recent work on Security Games with Signaling (SGS). However, current methods to solve SGS do not scale well in terms of time or memory. We therefore propose a novel approach to SGS, which, for the first time in this domain, employs an Evolutionary Computation paradigm: EASGS. EASGS effectively searches the huge SGS solution space via suitable solution encoding in a chromosome and a specially-designed set of operators. The operators include three types of mutations, each focusing on a particular aspect of the SGS solution, optimized crossover and a local coverage improvement scheme (a memetic aspect of EASGS). We also introduce a new set of benchmark games, based on dense or locally-dense graphs that reflect real-world SGS settings. In the majority of 342 test game instances, EASGS outperforms state-of-the-art methods, including a reinforcement learning method, in terms of time scalability, nearly constant memory utilization, and quality of the returned defender's strategies (expected payoffs). © 2022 International Joint Conferences on Artificial Intelligence. All rights reserved.

关键词： Chromosomes

来源：评论

学校读者我要写书评

暂无评论

Designing Robust Deep Learning Models for RealTime data Analysis in Machine Learning 3

Designing Robust Deep Learning Models for RealTime Data Anal...

引用

3rd International Conference on Smart Generation Computing, Communication and Networking, SMART GENCON 2023

作者： Borade, Monali R. Gupta, Mukur Faiz, Ayaan Acharjya, Kalyan Rajesh, A. Nunciya, P.M. Vishwakarma Institute of Information Technology Department of Artificial Intelligence & Data Science Pune India Vivekananda Global University Department of Electrical Engineering Jaipur India Centre of Interdisciplinary Research in Business and Technology Chitkara University Institute of Engineering and Technology Chitkara University Punjab India Maharishi School of Engineering and Technology Maharishi University of Information Technology Uttar Pradesh India Department of Computer Science Engineering Faculty of Engineering and Technology Karnataka Ramnagar562112 India Prince Shri Venkateshwara Padmavathy Engineering College Department of Master of Business Administration Chennai127 India

ISBN: (纸本)9798350319125

The developing call for real-time statistics analysis and selection-making has been a primary driver of improving deep studying models for the device getting to know. These models permit the evaluation of massive statistics sets, the generation of predictions from the statistics, and the speedy studying of new behaviors and patterns. Deep mastering fashions have been efficaciously used for various tasks, including natural language processing, computer imagination and prescient, and reinforcement studying. However, because of their complexity and a massive variety of parameters, they can be unreliable and volatile when uncovered to real-global records. This paper outlines strategies and techniques to lay out deep mastering fashions that can be robust to actual-world records. We also provide high-quality practices for deploying and retaining deep learning models in manufacturing packages. Those strategies include regularization techniques, model architectures, facts pre-processing techniques, and model validation and monitoring techniques. We additionally speak of the numerous strategies for assessing and improving the reliability of deep getting-to-know fashions. Subsequently, we provide a case study of a product that uses deep studying fashions to stumble on and classify objects in near-actual time in a manufacturing place. © 2023 IEEE.

关键词： improving maintaining production regularization techniques

来源：评论

学校读者我要写书评

暂无评论

A Sketch-Based Generation Model for Diverse Ceramic Tile Images Using Generative Adversarial Network

引用

Intelligent Automation & Soft Computing 2023年第9期37卷 2865-2882页

作者： Jianfeng Lu Xinyi Liu Mengtao Shi Chen Cui Mahmoud Emam School of Computer Science and Technology Hangzhou Dianzi UniversityHangzhou310018China Key Laboratory of Public Security Information Application Based on Big-Data Architecture Ministry of Public SecurityZhejiang Police CollegeHangzhou310000China Faculty of Artificial Intelligence Menoufia UniversityShebin El-Koom32511Egypt

Ceramic tiles are one of the most indispensable materials for interior *** ceramic patterns can’t match the design requirements in terms of diversity and interactivity due to their natural *** this paper,we propose a sketch-based generation method for generating diverse ceramic tile images based on a hand-drawn sketches using Generative Adversarial Network(GAN).The generated tile images can be tailored to meet the specific needs of the user for the tile *** proposed method consists of four ***,a dataset of ceramic tile images with diverse distributions is created and then pre-trained based on ***,for each ceramic tile image in the dataset,the corresponding sketch image is generated and then the mapping relationship between the images is trained based on a sketch extraction network using ResNet Block and jump connection to improve the quality of the generated ***,the sketch style is redefined according to the characteristics of the ceramic tile images and then double cross-domain adversarial loss functions are employed to guide the ceramic tile generation network for fitting in the direction of the sketch style and to improve the training ***,we apply hidden space perturbation and interpolation for further enriching the output textures style and satisfying the concept of“one style with multiple faces”.We conduct the training process of the proposed generation network on 2583 ceramic tile images *** measure the generative diversity and quality,we use Frechet Inception Distance(FID)and Blind/Referenceless Image Spatial Quality Evaluator(BRISQUE)*** experimental results prove that the proposed model greatly enhances the generation results of the ceramic tile images,with FID of 32.47 and BRISQUE of 28.44.

关键词： Ceramic tile pattern design cross-domain learning deep learning GAN generative adversarial networks ResNet Block

来源：评论

学校读者我要写书评

暂无评论

Drone Gesture Control using OpenCV and Tello

Drone Gesture Control using OpenCV and Tello

引用

Conference on information Communication technology and Society (ICTAS)

作者： Nokuthula Khoza Pius Owolawi Vusumuzi Malele Department of Computer Systems Engineering Faculty of Information and Communication Technology Soshanguve Campus Tshwane University of Technology Pretoria South Africa Unit for Data Science and Computing School of Computer Science and Information Systems Vanderbijlpark Campus North-West University Vanderbijlpark South Africa

Tracking a person with an onboard camera is a very difficult and perhaps technically impossible if one camera is used. In this regard, real-life projects use a series of cameras to achieve the task. The advent of camera installed drone technology has made it possible for researchers and designers to resolve the tracking issues of a single or multiple stagnant devices. This paper presents a method that makes it possible for a drone to recognize hand gestures as control mechanism and follow the person who control it. Furthermore, the method assists in drone object detection using an integrated depth camera that provides video capturing. The OpenCV and Tello were used in this project. A stabilized depth video is for use with current person trackers like the OpenCV tracker. The estimated Tello pose deduced from vision to stabilize the depth image by warping it to a virtual-static camera. This method allows the Tello to obtain not only the tracked person's position and orientation but also their full body pose, which can then be used, for instance, to recognize hand gestures and alter the Tello's behavior. This work has a potential of contributing or enhancing cutting-edge applications such as hands-free photography and videography.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel Image Encryption and Decoding Approach Based on DNA Coding and Arnold's Transformation for IoT Applications

A Novel Image Encryption and Decoding Approach Based on DNA ...

引用

Telecommunications Conference (ITC-Egypt), International

作者： Nawal Shaltout Ahmed A. Abd El-Latif Waleed Al-Adrousy Mohamed Hammad Abdelhamied A. Ateya Samir Elmougy Department of Computer Science Faculty of Computers and Information Mansoura University Mansoura Egypt EIAS Data Science Lab College of Computer and Information Sciences Prince Sultan University Riyadh Saudi Arabia Department of Mathematics and Computer Science Faculty of Science Menoufia University Egypt Department of Information Technology Faculty of Computers and Information Menoufia University Shibin El Kom Egypt Department of Electronics and Communications Engineering Zagazig University Zagazig Egypt

ISBN: (数字)9798350351408

ISBN: (纸本)9798350351415

Nowadays, the Internet of Things (IoT) is significantly utilized by an extensive range of applications that lead to the enhancement of different aspects of defense and military capabilities. In IoT, a wide range of smart devices that can communicate with each other are connected through it. Therefore, it is essential to protect multimedia data, especially images with an encryption approach that offers a sufficient degree of security in military applications. Hence, this paper introduces an effective scheme for image encryption and decoding relying on Arnold's transformation DNA coding. It proposes a new pseudo-random number generation PRNG using the Sine-Logistic map and Duffing map called SLDM. The encryption method involves three phases: 1- Key generation and PRNG. 2- confusion phase using the Arnold transformation method. 3- diffusion using DNA rules and operations. The results based on the USC-SIPI dataset, show that the suggested scheme is sufficiently resistant to every kind of attack. Furthermore, the suggested method yields NPCR scores of around 99.62% and UACI scores of approximately 33.48%. Due to its low complexity, it has a great deal of potential for safe and real-time picture applications.

关键词： Resistance Image coding Chaotic communication DNA Encryption Decoding Telecommunications

来源：评论

学校读者我要写书评

暂无评论

An Evaluation Metric for Single Image-to-3D Models Based on a Class Confidence Score of Object Detection Models

An Evaluation Metric for Single Image-to-3D Models Based on ...

引用

IEEE Global Conference on Consumer Electronics (GCCE)

作者： Yuiko Uchida Ren Togo Keisuke Maeda Takahiro Ogawa Miki Haseyama School of Engineering Hokkaido University Japan Faculty of Information Science and Technology Hokkaido University Japan Data-Driven Interdisciplinary Research Emergence Department Hokkaido University Japan

ISBN: (数字)9798350355079

ISBN: (纸本)9798350355086

This paper proposes a novel evaluation metric for single image-to-3D models based on a class confidence score of an object detection model. Traditional metrics, such as Structural Similarity Index and CLIP-similarity, often fail to accurately evaluate the quality of generated 3D objects. This is particularly evident when the objects have similar luminance but different meanings, or when the object is blurred. Our proposed metric can overcome these limitations by evaluating both semantic and visual consistency. Experimental results demonstrate that the proposed metric exhibits a higher rank correlation coefficient with the order of the model publication year compared to conventional metrics.

关键词： Measurement Correlation coefficient Visualization Three-dimensional displays Accuracy Semantics Object detection Indexes Consumer electronics

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：