检索结果-内蒙古大学图书馆

16th International Conference on Information technology and Applications, ICITA 2022

作者： Khalid, Fatima Javed, Ali Irtaza, Aun Malik, Khalid Mahmood Department of Computer Science University of Engineering and Technology Taxila Pakistan Department of Software Engineering University of Engineering and Technology Taxila Pakistan Department of Computer Science and Engineering Oakland University RochesterMI United States

ISBN: (纸本)9789811993305

In recent years, we have witnessed a tremendous evolution in generative adversarial networks resulting in the creation of much realistic fake multimedia content termed deepfakes. The deepfakes are created by superimposing one person’s real facial features, expressions, or lip movements onto another one. Apart from the benefits of deepfakes, it has been largely misused to propagate disinformation about influential persons like celebrities, politicians, etc. Since the deepfakes are created using different generative algorithms and involve much realism, thus it is a challenging task to detect them. Existing deepfakes detection methods have shown lower performance on forged videos that are generated using different algorithms, as well as videos that are of low resolution, compressed, or computationally more complex. To counter these issues, we propose a novel fused truncated DenseNet121 model for deepfakes videos detection. We employ transfer learning to reduce the resources and improve effectiveness, truncation to reduce the parameters and model size, and feature fusion to strengthen the representation by capturing more distinct traits of the input video. Our fused truncated DenseNet model lowers the DenseNet121 parameters count from 8.5 to 0.5 million. This makes our model more effective and lightweight that can be deployed in portable devices for real-time deepfakes detection. Our proposed model can reliably detect various types of deepfakes as well as deepfakes of different generative methods. We evaluated our model on two diverse datasets: a large-scale FaceForensics (FF)++ dataset and the World Leaders (WL) dataset. Our model achieves a remarkable accuracy of 99.03% on the WL dataset and 87.76% on the FF++ which shows the effectiveness of our method for deepfakes detection. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence Based Dynamic Approach to Visualize the Graphs 1

Artificial Intelligence Based Dynamic Approach to Visualize ...

引用

1st International Conference on Computational Intelligence, Communication technology and Networking, CICTN 2023

作者： Garg, Harshita Gupta, Ishika Kumar, Kapil Kaur, Baljeet Pundir, Divya Meerut Institute of Engineering & Technology Department of Computer Science and Engineering Meerut India

ISBN: (纸本)9798350338027

Understanding graph algorithms for beginners is the biggest nightmare for them. To conquer that fear we have decided to build a website which will help them to understand graph algorithms in a fun and easy way. It is a visualizer website for Graph Algorithms, where you can see a detailed explanation and animations of important algorithm runtime and build intuition. Graph Algorithm Visualizer allows you to easily construct graphs and step-by-step observe execution of algorithms on them. It visualizes different types of graph algorithms such as Prim's Algorithm, Kruskal Algorithm, DFS etc. Our aim is to build a project which will work for a teaching and learning tool. In learning mode, we are providing a detailed visualization of graph-based algorithms, such as Kruskal, Prim's algorithm as well as each step will be verified against the algorithm and results will be provided at a time. We have confidence that our project will help many students to understand graph-based algorithms at a deeper level and improve their knowledge of graph algorithms. This model is especially built to help faculty to teach graph algorithms in a creative and fun way for students. © 2023 IEEE.

关键词： Visualization

来源：评论

学校读者我要写书评

暂无评论

Forecast the Onset of Parkinson's Disease at all Three Stages using Deep Learning Techniques 1

Forecast the Onset of Parkinson's Disease at all Three Stage...

引用

1st International Conference on Advances in Electrical, Electronics and Computational Intelligence, ICAEECI 2023

作者： Caroline El Fiorenza, J. Sellam, V. Rose, J Dafni SRM Institute of Science and Technology Department of Computer Science and Engineering Chennai India St. Joseph's Institute of Technology OMR Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350342796

The most common neurological ailment & the 2nd most frequent reason for disability & mortality is Parkinson's disease (PD). A difficult challenge for making sure that people can live as independently as feasible is the correct identification of PD in the initial stages. Over the past few decades, artificial intelligence-based machine learning techniques for disease identification in medicine have become very popular. These techniques do not, however, offer a precise and rapid diagnosis. The total detection precision of methods connected to machine learning was insufficient. Italian medical control individuals and PD sufferers, mid-advanced patients treated to L-Dopa, separated into non-medication patients and early-diagnosed, were used to create a custom dataset using high-quality recordings of vocal assignments. The goal is to construct an accurate model that can accurately diagnose to identify PD early and begin treatment while it is still treatable, early diagnostic tools are required. However, current models contain drawbacks that can cause the disease to be misdiagnosed. The Unified PDRating Scale (UPDRS), a grading system for PD, could be continuously monitored with the use of the proposed framework, which could aid clinicians treat patients. According to the various ML pipelines, state-of-the-art is contrasted using various categorization methods&DL was also investigated utilizing a unique Convolutional Neural Network (CNN) framework. Outcomes demonstrate that KNN, SVM, and Naive Bayes classifiers operate comparably to each other with KNN having a little advantage over SVM and DL in terms of categorization outcomes. The dominance of CFS as the top characteristic choice is much clearer. The chosen characteristics serve as pertinent vocal biomarkers that can distinguish between healthy people, early-stage PD patients who have not yet received treatment, and mid-advanced PD patients who have received L-Dopa. The proposed approach has a 0.10 RMSE error, which is

关键词： Deep Learning Machine Learning Parkinson Disease Performance Measures Vocal biomarkers

来源：评论

学校读者我要写书评

暂无评论

Exploring Social Virtual Worlds for Virtual Communication 2

Exploring Social Virtual Worlds for Virtual Communication

引用

2nd IEEE International Conference on Vision Towards Emerging Trends in Communication and Networking Technologies, ViTECoN 2023

作者： Levin McLeish, M. Lakshmanan, M. Judgi, T. Sathyabama Institute of Science and Technology Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350347982

Social Virtual Worlds have begun to offer great potential for communication in recent years. The recent development of SVWs, VR and blockchain has led to the metaverse. SVW is all about creating an immersive virtual social environment that simulates a three dimensional social environment for communication, which is primarily based on personal presence and social presence. In this research we connect users in a surreal virtual environment and establish voice communication in order to make them feel each other's presence. Users can interact and hangout in the virtual world. © 2023 IEEE.

关键词： Virtual reality

来源：评论

学校读者我要写书评

暂无评论

Reliability Analysis of Network Connectivity on Wireless Ad Hoc Networks 14

Reliability Analysis of Network Connectivity on Wireless Ad ...

引用

14th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2023

作者： Mishra, Pranati Das, Ananya Department of Computer Science and Engineering Odisha University of Technology and Engineering Bhubaneswar India

Reliability on wireless ad hoc networks will offer a fault-tolerant network and identify faults that occur at various levels. The key task of the sensor is to monitor the area by gathering data and transmitting it to the destination node. Sensors failed as a result of their low cost and low energy capacity. Reliability requirements and disadvantages depend on the resources used by the wireless network. The paper addresses the issue of wireless ad hoc and sensor network reliability using a node and link failure model. The number of neighboring nodes, rate of the packet success, type of device, and the size of the packet are all factors that go into calculating node reliability. The reliability of a link among any node pair is estimated by taking into consideration the distance between nodes and the ratio of signal to noise . This method is projected to evaluate and calculate the reliability of wireless ad hoc networks, taking into consideration numerous routes from starting and destination nodes. Analyzed and discussed are the effects of various factors on reliability analysis of node and link. © Grenze Scientific Society, 2023.

关键词： Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

Identification of Personality Traits from Bangla Social Media Texts and Explaining with LIME 26

Identification of Personality Traits from Bangla Social Medi...

引用

26th International Conference on computer and Information technology, ICCIT 2023

作者： Islam, Ashiqul Arnob, Shakil Shahriar Hossain Ahsan, Younus Nahian, Sinan Muhammad Tabassum Tanny, Nawshin Ahsanullah University of Science and Technology Department of Computer Science and Engineering Dhaka Bangladesh

ISBN: (纸本)9798350359015

The extensive use of social media has provided people with a digital platform to express their ideas, sentiments, emotions, and opinions. This study focuses on the identification of personality traits from Bangla social media texts. Leveraging a diverse range of machine learning and deep learning methods, including BERT-based architectures, we improved the classification accuracy in this domain. The large variant of BanglaBERT emerged as the top performer, achieving an accuracy of 47.97%. While notable progress was made, challenges remain due to the dataset's size and the complexity of social media texts. © 2023 IEEE.

关键词： Bangla BERT Big Five OCEAN Personality traits Social media

来源：评论

学校读者我要写书评

暂无评论

An Intelligent Optimization Method of Reinforcing Bar Cutting for Construction Site

引用

computer Modeling in engineering & sciences 2023年第1期134卷 637-655页

作者： Zhaoxi Ma Qin Zhao Tianyou Cang Zongjian Li Yiyun Zhu Xinhong Hei Department of Civil Engineering and Architecture Xi’an University of TechnologyXi’an710048China State Key Laboratory of Rail Transit Engineering Informatization(FSDI) Xi’an710043China Department of Computer Science and Engineering Xi’an University of TechnologyXi’an710048China

To meet the requirements of specifications,intelligent optimization of steel bar blanking can improve resource utilization and promote the intelligent development of sustainable *** one of the most important building materials in construction engineering,reinforcing bars(rebar)account for more than 30%of the cost in civil engineering.A significant amount of cutting waste is generated during the construction *** cutting waste increases construction costs and generates a considerable amount of CO_(2)*** study aimed to develop an optimization algorithm for steel bar blanking that can be used in the intelligent optimization of steel bar engineering to realize sustainable *** the proposed algorithm,the integer linear programming algorithm was applied to solve the *** was combined with the statistical method,a greedy strategy was introduced,and a method for determining the dynamic critical threshold was developed to ensure the accuracy of large-scale data *** proposed algorithm was verified through a case study;the results confirmed that the rebar loss rate of the proposed method was reduced by 9.124%compared with that of traditional distributed processing of steel bars,reducing CO_(2)emissions and saving construction *** the scale of a project increases,the calculation quality of the optimization algorithmfor steel bar blanking proposed also increases,while maintaining high calculation *** the results of this study are applied in practice,they can be used as a sustainable foundation for building informatization and intelligent development.

关键词： Building construction rebar work cutting stock problem optimization algorithm integer linear programming

来源：评论

学校读者我要写书评

暂无评论

Adapting to Evasive Tactics through Resilient Adversarial Machine Learning for Malware Detection 18

Adapting to Evasive Tactics through Resilient Adversarial Ma...

引用

18th INDIAcom;11th International Conference on Computing for Sustainable Global Development, INDIACom 2024

作者： Krishna, G. Bala Sravan Kumar, G. Ramachandra, Mummadi Sampurnima Pattem, K. Rani, D. Sandhya Kakarla, Geeta Cvr College of Engineering Department of Computer Science Engg. Rangareddy Ibrahimpatnam India Cvr College of Engineering Department of Computer Science Engg.-Cyber Security Rangareddy Ibrahimpatnam India Sreenidhi Institute of Science and Technology Department of Computer Science Engg. Hyderabad India Sreenidhi Institute of Science and Technology Department of Computer Science Engg. Cyber Security Hyderabad India

ISBN: (纸本)9789380544519

This paper presents the Adaptive Resilience-based Convolutional Network (ARCNet), a sophisticated machine learning framework specifically designed to detect advanced, evasive malware. ARCNet combines convolutional and recurrent neural networks, making it highly adaptable to changing cyber threats. Its core components, the Adversarial Learning Module (ALM), Predictive Analysis Engine (PAE), and Dynamic Adaptation System (DAS), significantly boost its detection power. Tests using a synthetic dataset show ARCNet's superiority over traditional models like the Support Vector Machine (SVM). It achieved 95.2% accuracy under normal conditions (compared to SVM's 89.4%) and maintained 92.5% accuracy even during adversarial attacks (against SVM's 80.3%). Notably, ARCNet's detection rates improved from 78.5% to 86.7% in five months after integrating the DAS. These results confirm ARCNet's efficiency in tackling complex malware challenges, contributing greatly to cybersecurity. The study underscores the importance of evolving and enhancing machine learning methods to keep pace with the rapidly changing landscape of cyber threats. © 2024 Bharati Vidyapeeth, New Delhi.

关键词： Block chain Blockchains Challenges of Block chain Consensus algorithm Data privacy Digital signatures Distributed ledger Faces Organizations Smart contracts

来源：评论

学校读者我要写书评

暂无评论

A Reputation-Based AODV Protocol for Blackhole and Malfunction Nodes Detection and Avoidance

引用

computers, Materials & Continua 2024年第8期80卷 1867-1888页

作者： Qussai M.Yaseen Monther Aldwairi Ahmad Manasrah Artificial Intelligence Research Center(AIRC) College of Engineering and Information TechnologyAjman UniversityAjman346United Arab Emirates Department of Computer Information Systems Faculty of Computer and Information TechnologyJordan University of Science and TechnologyIrbid22110Jordan College of Technological Innovation Zayed UniversityAbu Dhabi144534United Arab Emirates Faculty of Information Technology and Computer Science Yarmouk UniversityIrbid21163Jordan Computer Information Systems Higher Colleges of TechnologySharjah341214United Arab Emirates

Enhancing the security of Wireless Sensor Networks(WSNs)improves the usability of their ***,finding solutions to various attacks,such as the blackhole attack,is crucial for the success of WSN *** paper proposes an enhanced version of the AODV(Ad Hoc On-Demand Distance Vector)protocol capable of detecting blackholes and malfunctioning benign nodes in WSNs,thereby avoiding them when delivering *** proposed version employs a network-based reputation system to select the best and most secure path to a *** achieve this goal,the proposed version utilizes the Watchdogs/Pathrater mechanisms in AODV to gather and broadcast reputations to all network nodes to build the network-based reputation *** minimize the network overhead of the proposed approach,the paper uses reputation aggregator nodes only for forwarding reputation ***,to reduce the overhead of updating reputation tables,the paper proposes three mechanisms,which are the prompt broadcast,the regular broadcast,and the light broadcast *** proposed enhanced version has been designed to perform effectively in dynamic environments such as mobile WSNs where nodes,including blackholes,move continuously,which is considered a challenge for other *** the proposed enhanced protocol,a node evaluates the security of different routes to a destination and can select the most secure routing *** paper provides an algorithm that explains the proposed protocol in detail and demonstrates a case study that shows the operations of calculating and updating reputation values when nodes move across different ***,the paper discusses the proposed approach’s overhead analysis to prove the proposed enhancement’s correctness and applicability.

关键词： AODV blackhole malfunction nodes pathrater reputation system wireless sensor networks(WSNs) watchdogs

来源：评论

学校读者我要写书评

暂无评论

Air Canvas: Expressing Creativity in the Real World with OpenCV and Python

Air Canvas: Expressing Creativity in the Real World with Ope...

引用

2023 International Conference on Energy, Materials and Communication engineering, ICEMCE 2023

作者： Shanthalaksmi Revathy, J. Suresh, A. Sanjith Bala, B. Velammal College of Engineering and Technology Department of Computer Science and Engineering Madurai India

ISBN: (纸本)9798350393378

In today's world, online education has become one of the most important and widely used platform as it has spread all over the world. But online learning also has many challenges. One of the most important problems is to focus on the courses taught. To solve this problem, we created an "air canvas"that allows content to be delivered in real space and allows us to draw our needs by holding our fingers in the air.. It works by using a webcam to capture finger movements. It is built using the famous OpenCV library and python. There are numerous ways to write when drawing, including utilising a keyboard, touch screen, digital pen, pencil, or electric glove. But in the proposed system, Python computer vision programming and gesture detection are used to produce natural human-computer interaction. © 2023 IEEE.

关键词： Python

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：