检索结果-内蒙古大学图书馆

2024 International Conference on Artificial Intelligence and Emerging technology, Global AI Summit 2024

作者： Ghai, Kapil Anupriya Jain, Paras Gupta, Vishan Kumar Vidisha Shakya, Harish Graphic Era hill University Department of Chemistry UK Dehradun India School of Computing Graphic Era Hill University UK Dehradun India School of Computing Engineering and Artificial Intelligence Vit Bhopal University Kothri Kalan MP Bhopal India Amity School of Engineering and Technology Amity University Punjab Mohali India Manipal University Department of Computer Science and Engineering Rajasthan Jaipur India

ISBN: (纸本)9798350379716

Emotion recognition as the aspect of artificial intelligence is the ability of a system to understand people's emotions from information that includes face, voice, and text. More complex algorithms such as CNNs and RNNs are used to train the algorithms on considerable amounts of data to extract relevant features for high reliability and efficiency. Measures such as accuracy, and preciseness of the model often determine the usability of the models in practical applications. The proper implementation of these models has major repercussions in various domains, including human computer interaction, healthcare and sentiment analysis to name a few, changing the idea and approach towards emotions in the modern world. These approaches offer improved accuracy and robustness, setting the stage for more nuanced and reliable emotion recognition systems. We have gathered dataset for emotion recognition from Kaggle. We used a dataset that contains various features extracted from various emotions images, and our model resulted in a test accuracy of 95 %. This high level of accuracy demonstrates the capability of our model to effectively classify emotional states. © 2024 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Quantum Man-in-the-Middle Attacks on QKD Protocols: Proposal of a Novel Attack Strategy 6

Quantum Man-in-the-Middle Attacks on QKD Protocols: Proposal...

引用

6th International Conference on Contemporary Computing and Informatics, IC3I 2023

作者： Amellal, Hicham Hajjami, Salma El Kaushik, Keshav Chhabra, Gunjan Yadav, Suman Avdhesh LabSIV Faculty of Science Department of Computer Science Agadir Morocco School of Computer Sciences University of Petroleum and Energy Studies Uttarakhand Dehradun India Graphic Era Hill University Department of CSE Uttarakhand Dehradun India Amity School of Engineering and Technology Department of Computer Science Amity University Uttar Pradesh Greater Noida India

ISBN: (纸本)9798350304480

Quantum Key Distribution (QKD) protocols theo- retically propose a high-security level based on the advantages of quantum computing. This paper analyze the security level of one of the most well-known QKD protocols, BB84, against the man-in-the-middle attack strategy. Specifically, we leverage quantum entanglement to enhance communication security by the injection of a malicious photon in the middle of transmission. In the security analysis, we measure the variation in mutual information between the sender-receiver and sender-attacker to assess the protocol's resilience against potential attacks. © 2023 IEEE.

关键词： BB84 Protocol man-in-the-middle attack QKD protocols security

来源：评论

学校读者我要写书评

暂无评论

An Analysis of Machine Learning Methods used in Cyber Security during the Last Decade

An Analysis of Machine Learning Methods used in Cyber Securi...

引用

2024 International Conference on Intelligent and Innovative Practices in engineering and Management, IIPEM 2024

作者： Suresh Babu, Rakesh Thoppaen Dhar Diwan, Tarun Vishnoi, Ankit William, P. Madan, Parul Kaushik, Keshav HEXAWARE TECHNOLOGIES INC 101 Wood Avenue South IselinNJ08830 United States Atal Bihari Vajpayee University Bilaspur India Graphic Era Deemed to Be University Department of Computer Science and Engineering Uttarakhand Dehradun India Sanjivani College of Engineering Deaprtment of Information Technology Kopargaon India Amity University Dubai United Arab Emirates Amity University Punjab Amity School of Engineering and Technology Mohali India

ISBN: (纸本)9798350390049

This paper examines cybersecurity machine learning (ML) methods over the last decade. In this study, we examine how cyberthreats have evolved and how machine learning has progressed to combat them. Each problem is well discussed. Both topics are being discussed extensively. Cybersecurity has long employed several machine learning methods. This category includes reinforcement, supervised, and unsupervised learning methods. Other teaching methods include unsupervised learning. These methods excel in identifying possible threats. This article provides an overview of machine learning (ML)-based cybersecurity solutions to highlight major topics, issues, and research opportunities. The study uses academic, commercial, and real-world data to do this. It highlights important issues, difficulties, and opportunities. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Towards Optimized Physical Activity: A Fusion of Graph Convolutional Networks and XGBoost for Exercise Analysis

Towards Optimized Physical Activity: A Fusion of Graph Convo...

引用

Smart Systems for Electrical, Electronics, Communication and computer engineering (ICSSEECC), International Conference on

作者： Kishan V Rajesh M Dhanush B Rajiv Vincent School of Computer Science and Engineering Student Vellore Institute of Technology Chennai Chennai India School of Computer Science and Engineering Vellore Institute of Technology Chennai Chennai India School of Computer Science and Engineering Student Vellore Institute of Technology Chennai Chennai India

ISBN: (数字)9798350378177

ISBN: (纸本)9798350378184

This research explores the implementation of the GCN-XGBoost with attention mechanism within the detection and correction of exercises through computer vision, utilizing the 2-dimensional pose coordinates from the MMFit [ 1 ] dataset. The objective is to enhance the precision and reliability of squat performance analysis, leveraging the robust, scalable machine learning capabilities of XGBoost. By processing and analyzing two-dimensional skeletal data, the study aims to identify key postural deviations and provide corrective feedback in realtime, potentially reducing the risk of injury and improving overall exercise effectiveness. Initial results demonstrate the model's capability to discern between correct and incorrect exercise forms with high accuracy, indicating the feasibility of applying advanced machine learning techniques to personal fitness and rehabilitation contexts. This paper details the methodology, from data preprocessing to model training and evaluation, highlighting the contributions to the fields of sports science and machine learning in physical activity monitoring. We have used MMFit dataset for the model that we have trained. We use the 2D coordinates of 18 of the 33 joints in the human body as the main features to categorize various exercises and forecast the accuracy of the poses identified by the Google's AI framework MediaPipe [2]. In this paper, we first classify an individual's exercises (squats, lunges, bicep curls, situps, pushups, triceps extensions, dumbbell rows, jumping jacks, dumbbell shoulder press, lateral shoulder raises) and check whether the pose associated with an exercise is correct. The final model trained has a 98% final validation accuracy.

关键词： Training Presses Accuracy Lung Internet Performance analysis Reliability

来源：评论

学校读者我要写书评

暂无评论

Impact of COVID-19 Variants Using Multi-Criteria Selection Approach

Impact of COVID-19 Variants Using Multi-Criteria Selection A...

引用

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

作者： Shrddha Sagar Avadhesh Kumar Sofia Singh Garima Shukla Diwakar Upadhyay School of Computer Science and Engineering Galgotias University Uttar Pradesh India Department of Artificial Intelligence Amity School of Engineering and Technology Amity University Noida Uttar Pradesh Department of Computer Science and Engineering Amity School of Engineering and Technology Amity University Mumbai Maharashtra School of Computing Graphic Era Hill University Haldwani India

ISBN: (数字)9798331517953

ISBN: (纸本)9798331517960

The aim of the proposed work is related to the detailed study of various types of COVID-19 along with its symptoms. This study utilized a retrospective approach to the COVID data from 2019 to 2021 to assess the severity level of the various types of COVID-19. After a detailed analysis, we have identified seven symptoms which are as follows: breathing, chest pain, fever, body ache, sore throat and cough, tiredness, and loss of taste/smell. In this study, the prediction of the COVID-19 severity level is made using a fuzhzy multi-objective optimization based on ratio analysis (fuzzy MOORA) approach. The proposed model simultaneously predicts the severity level of COVID-19 with multiple criteria. The severity levels listed on government-approved websites for various types of COVID-19 are considered for validating the proposed methodology's result. The highest overall performance index value is used to decide. As a result, the overall performance index value has been achieved and moreover, we can predict the severity level of different forms of COVID-19 using this strategy. After the implementation of our proposed approach, we got the overall performance index value of delta T4 (0.433086). The Coronavirus Virus 2019 (COVID-19) pandemic is on the increase for over two years. The World Health Organization (WHO) statistics show that approximately 260 million confirmed cases have been documented, with over five million deaths. In our work, we have implemented a fuzzy MOORA approach for predicting the severity level of the different types of COVID-19. The overall performance index value of each COVID-19 is shown in the table. Table 1.2 demonstrates that Delta T4 is ranked 1 st , Gamma T3 is ranked 2 nd , Beta T2 is ranked 3 rd , Alpha T1 4th and Omicron T5 has the least rank.

关键词： COVID-19 Quantum computing Pandemics Pain computer viruses Organizations Predictive models Performance analysis Artificial intelligence Optimization

来源：评论

学校读者我要写书评

暂无评论

Enabling Speech Recognition for Lesser-Known Language 6

Enabling Speech Recognition for Lesser-Known Language

引用

6th International Conference on Contemporary Computing and Informatics, IC3I 2023

作者： Jha, Amit Kushwaha, Pradeep Kumar Srivastava, Arun Pratap Thakur, Anil Kumar, Devesh Gupta, Shubhi Amity School of Engineering and Technology Department of Computer Science Uttar Pradesh India Lloyed Institute of Engineering and Technology Uttar Pradesh Greater Noida India Lloyd Law College Uttar Pradesh Greater Noida India Amity University Uttar Pradesh Greater Noida India

ISBN: (纸本)9798350304480

An overview of Automatic Speech Recognition (ASR) technology, including its history and challenges in implementing it for regional languages. The challenges include lack of training data, dialectal and pronunciation variations, limited vocabulary, code mixing, and lack of standardization. Addressing these challenges requires significant research and development efforts to create accurate and reliable ASR systems for regional languages. In this research paper we have discussed various Automatic Speech Recognition model along with the application used in various field finally we have evaluated the accuracy of ASR for major Indian languages. © 2023 IEEE.

关键词： ASR Deep Neural Network Hidden Markov Model

来源：评论

学校读者我要写书评

暂无评论

Framework to analyze and exploit the smart home IoT firmware

引用

Measurement: Sensors 2025年 37卷

作者： Kaushik, Keshav Bhardwaj, Akashdeep Dahiya, Susheela School of Computer Science University of Petroleum and Energy Studies Uttarakhand Dehradun India Amity School of Engineering and Technology Amity University Punjab Mohali India Department of Computer Science and Engineering Graphic Era Hill University Uttarakhand Dehradun India

Thousands of smart gadgets are linked to the internet each month, and due to various privacy and security issues, such devices may be vulnerable to evil attackers. Currently, there are around 8 billion devices online, and by the beginning of 2025, there will likely be 25 to 35 billion IoT devices globally. Despite this, the security of the devices is not receiving any more attention. Since operating system (OS) and hardware security have improved recently, researchers and hackers now seek vulnerabilities in other areas, such as firmware. When the firmware on many IoT devices isn't updated, it leaves them open to cyberattacks. The period when the Mirai Botnet was widely used is one of the situations in which we may have heard about firmware security. By gaining access to the equipment using the default credentials, the Mirai Botnet infects devices. Therefore, to analyze the firmware's contents for alteration during runtime, the authors of this research performed reverse engineering on it. Authors have exploited the smart home IoT firmware using our framework that identified ten critical network-based vulnerabilities within the firmware, with five vulnerabilities scoring a maximum CVSS score of 10.0 and the remaining five scoring 9.8, highlighting significant threats to smart home IoT devices. In addition, examining the firmware binaries demonstrates the widespread usage of dangerous functions like sprintf and strcpy in addition to the absence of critical security features like NX, PIE, RELRO, and stack protection. By offering a thorough analysis of the vulnerabilities and suggesting best practices for boosting the security of smart home IoT firmware, the results add to the body of information already in existence. © 2024 The Authors

关键词： Smart homes

来源：评论

学校读者我要写书评

暂无评论

A Proposed Framework for Robot-Mediated Healthcare and Response Systems

A Proposed Framework for Robot-Mediated Healthcare and Respo...

引用

International Conference on Inventive Research in Computing Applications (ICIRCA)

作者： Hafsah Khan Deepak Arora Sheenu Rizvi Department of Computer Science & Engineering Amity School of Engineering and Technology Amity University Lucknow India

The study proposes an experimental model for robot-mediated systems in healthcare. The framework outline and required modules are discussed successively alongside the methodology needed to achieve the system's base functioning. The concept of response systems and their importance in building an efficient agent for healthcare support are also discussed. The study looks at the concept of human-machine interfaces, medical robotic systems, methods of implementation and design to achieve the conception, the underlying effects, and the public's opinion on technology in healthcare. The proposed system derives from the issues faced in conveying the full capabilities of a healthcare network. The idea behind the proposal is to provide a means to leverage different components of healthcare systems to attain a holistic model capable of serving various patient and staff needs.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Online social network automation attack detection methods for energy analysis and consumption modelling

引用

International Journal of Information technology (Singapore) 2024年 1-13页

作者： Rawat, Romil Sikarwar, Ranjana Maravi, Prakash Kumar Ingle, Mandakini Bhardwaj, Vivek Rawat, Anjali Rawat, Hitesh Department of Computer Science Engineering Shri Vaishnav Institute of Information Technology Shri Vaishnav Vidyapeeth Vishwavidyalaya Indore Indore India Department of CSE Amity University Madhya Pradesh Gwalior India Department of Computer Science Engineering Parul University Gujarat Vadodara India Department of CSE Medicaps University Indore India School of Computer Science and Engineering Manipal University-Jaipur Rajasthan Jaipur India Department of Computer and Communication Technology University of Extremadura Badajoz Spain Department of Business Management and Economics University of Extremadura Badajoz Spain

Preventing the spread of Domain Name System (DNS) tunnelling and the use of automation infrastructure is one of the key problems that the idea of a contemporary smart OSN (online social network) must address. The proposed technique focuses on online social network automation attacks (OSNAA) based on smart social network devices (SSND) energy consumption (Ecomp) analysis based on user’s selection modes (USM) for the detection of OSNAA and Man-in-the-Middle (MiTM) attacks. Automated opcode sequence analysis (OSA) is used to improve the accuracy of OSNAA detection and localise MiTM attacks. It analyses the performance of the SSND for MiTM attacks by localising suspicious automated tools on these devices with applicable accuracy and is based on the monitoring of energy consumption, which helps to analyse and categorise the behaviour of the SSND in the state of normal or in compromised conditions by the threat agents. The suggested technique enables high-efficiency detection of SSND threats, such as MiTM attacks, at a level of about 99.91% and localization of suspicious automation tools on these machines with an accuracy of around 99.73%. © Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： Cyberattack Data mining Energy efficiency and consumption MiTM Opcode analysis OSN-Automation

来源：评论

学校读者我要写书评

暂无评论

Role of Smali Opcodes in Identifying Android Malware Obfuscation Variants

Role of Smali Opcodes in Identifying Android Malware Obfusca...

引用

International Conference on computer, Communication, Control and Information technology (C3IT)

作者： Abhishek Anand Jyoti Prakash Singh Amity School of Engineering and Technology Amity University Patna Patna India Computer Science and Engineering National Institute of Technology Patna Patna India

ISBN: (数字)9798350389128

ISBN: (纸本)9798350389135

With the widespread adoption of Android and its applications, obfuscated malware poses significant challenges to the platform's security. Addressing these challenges requires automated, lightweight, and scalable models capable of understanding obfuscation variants and distinguishing obfuscated malware from non-obfuscated malware. This paper proposes a lightweight model based on static analysis, utilizing reverse engineering techniques to extract Smali opcodes. For efficient, fast, and accurate analysis, we employ the SelectKBest feature selection method to identify the most important opcodes critical for recognizing obfuscation variants. The machine learning classifier Random Forest is used for the classification tasks, distinguishing between obfuscated and non-obfuscated malware and identifying specific obfuscation variants with high precision and accuracy. Our approach achieves an impressive accuracy of up to 99.32% in detecting obfuscation and an accuracy of 83.56% in classifying different obfuscation variants. This demonstrates the potential of our lightweight model to significantly enhance the security of the Android platform by effectively identifying and classifying obfuscated malware.

关键词： Analytical models Accuracy Operating systems Reverse engineering Static analysis Feature extraction Malware Security Information technology Random forests

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：