检索结果-内蒙古大学图书馆

Enhancing Educational Video Discovery Using Advanced Latent Semantic Analysis

Procedia computer science 2025年 252卷 784-795页

作者： B. Sindhu A. Bhaskar G. Yugesh S. Reshma B. Rohit Assistant Professor Department of Computer Science and Engineering Godavari Global University Rajamahendravaram -533296 Andhra Pradesh India

The rapid growth of online educational videos has created abundant learning resources, but also has challenges in discovering high-quality, relevant content. This project mainly focuses on video-based content analysis and providing high-quality learning sources for scholars, easing the best possible way to discover accurate resources. Existing Systems often rely on round-robin, random selection which fails to grasp semantics and linguistic complexity. The proposed system focuses on content-based analytics using NLP, where the extraction of transcripts from videos, perform relevance assessment which maps content to concept and extract insights using Latent Semantic Analysis which effectively captures underlying structures in the transcript. Complexity Assessment measures readability metrics using the Flesch-Kincaid Grade Level and SMOG Index. Created a web-based interface for easy access to educational videos, with built-in analytics to assess relevance. The insights are shared with scholars to refine content and better meet user requirements.

关键词： Latent Semantic Analysis NLP(Natural Language Processing) SMOG(Simple Measure of Gobbledygook) Index Topic Modelling TF-IDF (Term frequency-inverse document frequency) Content-based filtering

来源：评论

学校读者我要写书评

暂无评论

Framework for 2D MRI Brain Tumour Segmentation Employing Znet-Based Deep Learning 4th

Framework for 2D MRI Brain Tumour Segmentation Employing Zne...

引用

4th International Conference on Recent Trends in Machine Learning, IoT, Smart Cities, and Applications, ICMISC 2023

作者： Tejashwini, P.S. Thriveni, J. Venugopal, K.R. Bangalore Institute of Technology Bengaluru560004 India Professor Department of Computer Science and Engineering University of Visveshwaraya College of Engineering Bengaluru560001 India Former Vice Chancellor Bangalore University Karnataka Bengaluru India

ISBN: (纸本)9789819994410

The spotting and segmentation of brain tumours using MR images presents substantial problems to the medical community. For potentially life-saving actions and the ability for doctors to select efficient treatment modalities, early detection and accurate localization of brain tumours are essential. This research presents a unique structure called Znet that makes use of deep neural networks (DNN) and data augmentation methods with the goal to forecast and segment brain tumours in medical imaging. The architecture uses skip-connections, encoder–decoder models, and data amplification to create synthetic situations using the knowledge from a small number of expertly described tumours. Experimental results, which generated a high mean dice similarity coefficient (DSC) of about 0.98 as part of model training and 0.94 for the independent testing dataset, show the effectiveness of the Znet model. These results show how correct data estimation in the area of healthcare data can be improved by deep learning (DL). © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Tumors

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach to Prevent Hijacking of Accounts in the Cloud

A Novel Approach to Prevent Hijacking of Accounts in the Clo...

引用

computer science and Emerging Technologies (CSET), International Conference on

作者： Ananthoju Vijay Kumar Yeshitla Woinshet Mihret Nesru Kemal Nephtalem Woldeyohanes Ahmad Shah Dagemawi Ameha Computer Science and Engineering Associate Professor Jain University Bangalore India Computer science and engineering student Jain University Bangalore India

According to studies, majority of enterprises globally are now adopting cloud services, and the number of migrants to cloud environments in the industrialized nations is expanding ever since. Moving business data from a centrally hosted server to cloud storage has a lot of merits. However, it increases the chance of data security breach. An aaccount hijacking is the most frequent type of attack that happens in cloud system, it is a type of attack where hackers will get the account holder's credentials and control over the account. Considering the fact that cloud system is vulnerable to attacks, it is crucial to have a remediation mechanism like a secured authentication system. Authentication system which uses textual password can be used in cloud system, but it is vulnerable to regular attacks like brute-force attack, spyware attack, and so many more. The present research mainly focuses on creating a unique authentication system that prevents account hijacking in cloud by mitigating common attacks. To achieve the key objective a graphical password authentication system has been developed with OTP and colouring scheme in order to specifically minimize brute force and shoulder surfing attacks. It is a simple and easy to remember authentication system for customers while minimizing account hijacking in cloud at the same time. It is proved that the present approach is successful authentication system that is less vulnerable to shoulder surfing and brute force attack while involving easy way for customers to remember their passwords without hassle.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Blockchain-based Secure UAV-assisted Battlefield Operation underlying 5G 14

Blockchain-based Secure UAV-assisted Battlefield Operation u...

引用

14th International Conference on Information and Knowledge Technology, IKT 2023

作者： Pancholi, Dhruvi Jadav, Nilesh Kumar Tanwar, Sudeep Garg, Deepak Zanjani, S. Mohammadali Institute of Technology Nirma University Department of Computer Science and Engineering Ahmedabad 382481 India SR University Distinguished Professor School of Computer Science and Artificial Intelligence Warangal 506371 India Islamic Azad University Najafabad Branch Department of Electrical Engineering Najafabad Iran

ISBN: (纸本)9798350349412

The use of unmanned aerial vehicles (UAVs) has significantly increased during the past few years. UAVs have a wide range of applications, but their employment in the defense sector has grown significantly in recent years as a result of their exceptional effectiveness in reducing casualties and enabling the accomplishment of high-profile as well as time-sensitive operations. UAVs are helpful for numerous defense-related operations, such as border patrol, surveillance, emergency management, damage assessment, delivery of medical supplies and armaments, and many others. However, because wireless communication is open, UAVs can be used to undermine the effectiveness of military operations. In this paper, we developed a blockchain-based framework to mitigate security threats from UAVs operating in a combat scenario. The MATLAB virtual environment is used to simulate a battlefield scenario, including UAVs, sensors, enemies, and friendly nodes. The UAV scenario also includes an enemy that continuously manipulates the UAVs to endanger battlefield operations. The suggested framework employs machine learning algorithms to find malicious UAVs. Further, the proposed framework uses blockchain technology to increase security and lessen data tampering attempts. A smart contract (for data validation) is deployed on an IPFS-based public blockchain. In addition, 5G technology is used, which makes it possible for UAVs and the control station to communicate effectively with lower packet error rates and latency. The proposed framework is evaluated by considering evaluation metrics, such as 5G packet error rate, latency, and transaction and execution costs for the blockchain. © 2023 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Review of Segmentation Techniques for Weed Detection in Agricultural Crops

引用

Procedia computer science 2025年 259卷 1793-1802页

作者： Akanksha Bodhale Seema Verma Research Scholar Department of Computer Science & Engineering Banasthali Vidyapith Professor Department of Electronics School of Physical Sciences Banasthali Vidyapith

This study explores the role of deep learning in identifying and managing weeds in agriculture, a critical challenge for enhancing crop productivity. As the global population grows, increasing food production is essential. Weeds significantly hinder crop growth, making accurate identification vital. Deep learning techniques, which analyze elements like color, form, texture, and spectrum, offer promising solutions for distinguishing between crops and weeds. This review examines various segmentation techniques used in weed identification, comparing their effectiveness and potential for practical application. The findings aim to advance weed management strategies, contributing to improved agricultural productivity and the development of automated systems for precise weed detection.

关键词： Agriculture Weed Detection Segmentation Machine Learning Deep Learning

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of RPL Protocol Under Decreased and Increased Rank Attacks: A Focus on Smart Home Use-Case

引用

SN computer science 2023年第4期4卷 329页

作者： Bang, Ankur Rao, Udai Pratap Knowledge Associate Centre for Development of Advanced Computing (C-DAC) Maharashtra Mumbai 400049 India Computer Science and Engineering Department National Institute of Technology Bihar Patna 800005 India

The rank value of a node in Routing Protocol for Low Power and Lossy Networks (RPL) plays a significant role in preventing loops and route optimization. Advertising a fake rank value creates a false topology and can degrade network performance to a great extent. This manuscript aims to study the effects of increased and decreased variants of the rank attack on the RPL performance. We consider a smart home network topology to evaluate the RPL performance, and the simulation experiments are performed using the Cooja simulator. Through this study, we portray a rational analysis and discuss how the network performance can be degraded by each variant of the rank attack and the position of the attacker node in the network topology. The simulation results obtained in our work demonstrate how each variant of the rank attack can disrupt the default RPL working. We notice significant effects of both the studied rank attacks on the crucial parameters like packet overhead, average inter-packet time, packet delivery ratio and average power consumption. Researchers working in this field can take advantage of our research findings and extend their study to develop new defence mechanisms against these attacks. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： IoT Performance evaluation Rank attack RPL Smart home

来源：评论

学校读者我要写书评

暂无评论

Investigation of Security Attacks in IoMT Devices and Federated Learning as a Mitigation Strategy

引用

Procedia computer science 2025年 258卷 3426-3435页

作者： Kiruthika M T Poongodi Assistant Professor Department of Computational Intelligence School of Computing SRM Institute of Science and Technology Kattankulathur Tamilnadu 603203 India Professor Department of Computer Science and Engineering School of Engineering Dayananda Sagar University Bangalore Karnataka India

Internet of Medical Things (IoMT) is transforming the healthcare business with an increasing number of wearables, gadgets, and sensing devices. As a result, several patients are under daily monitoring for their cardio health, blood glucose level, blood pressure, etc. The data generated by these devices are very voluminous, and there is an increasing need for safeguarding these sensitive data in order to protect them from malicious attacks. This paper suggests to incorporate Federated learning model to IoMT as a mitigative measure. It is apt to induct FL into IoMT as it operates on a collaborative environment which do not allow transfer of the raw data. Patients’ data are stored at the client end itself and only model updates are shared to the server. In this way, the data and network attacks can be reduced to a great extent. This work proposes an integrated architecture where the layers of FL and IoMT can be knotted up for ensuring privacy and security. This proposed architecture model, further elaborates a distributed working environment. This paper illustrates how FL helps in securing the data produced by the different IoMT devices through the proposed architecture. The challenges involved in the deployment of FL in the framework of IoMT along with potential research directions are discussed further in this research.

关键词： : Federated Learning IoMT Security Internet of Medical Things IoMT Data Attacks Optimization

来源：评论

学校读者我要写书评

暂无评论

A novel stable feature selection algorithm for machine learning based intrusion detection system

引用

Procedia computer science 2025年 252卷 738-747页

作者： Sowmya T Mary Anita E A Department of Computer Science and Engineering Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal 576104 Karnataka India Research Scholar Department of Computer Science and Engineering School of Engineering and Technology Christ(Deemed to be University) Bangalore Professor Department of Computer Science and Engineering CHRIST (Deemed to be University) Bengaluru Karnataka 560074 India

The advent of new technologies like artificial intelligence, and big data has influenced many cyber attackers to launch their attacks on the network. Hence researchers have already proposed Intrusion Detection Systems by incorporating machine learning as well. Building an effective IDS is still a challenging task because of low accuracy. Managing high dimensional data is another major problem that occurs in IDS. Hence in this paper, an efficient Machine Learning based Intrusion Detection System is developed by means of a novel stable feature selection strategy called IV-RFE. The proposed methodology aims to select only the relevant features that contribute to the attack, which is purely based on relative variance, and weight factor in combination with RFE. This methodology increases the performance in terms of accuracy and maintains a stable set of features. Previous studies only focussed on the feature selection strategy and their performance. The feature stability also has to be considered which is an equally important metric, especially in the field of Intrusion Detection Systems. Hence in the current study, an efficient ML based IDS is proposed which selects only the relevant and stable features. Experimental results also revealed that the proposed IV-RFE outperformed well for three attacks with respect to accuracy and stability metrics also. The results show that stability is also an important indicator in selecting the features in the field of Intrusion Detection Systems.

关键词： Intrusion Detection System(IDS) Feature Selection(FS) Recursive Feature Elimination(RFE) Machine Learning

来源：评论

学校读者我要写书评

暂无评论

A Novel Technique for Detecting Various Thyroid Diseases Using Deep Learning

引用

Intelligent Automation & Soft Computing 2023年第1期35卷 199-214页

作者： Soma Prathibha Deepak Dahiya C.R.Rene Robin Cherukuru Venkata Nishkala S.Swedha Department of Information Technology Sri Sai Ram Engineering CollegeChennai600044India College of Computer and Information Sciences(CCIS) Majmaah UniversityMajmaah11952Kingdom of Saudi Arabia Department of Computer Science and Engineering Sri Sai Ram Engineering CollegeChennai602109India Associate Solutions Engineer PegasystemsChennai600044India Systems Engineer Infosys Ltd.Chennai600119India

Thyroid disease is a medical condition caused due to the excess release of thyroid *** is released by the thyroid gland which is in front of the neck just below the *** pictures such as X-rays and CT scans can,however,be used to diagnose *** this proposed model,Deep Learning technology is used to detect thyroid diseases.A Convolution Neural Network(CNN)based modified ResNet architecture is employed to detectfive different types of thyroid diseases namely *** *** *** cancer *** *** *** the proposed work,the training method is enhanced using dual optimizers for better accuracy and ***,a Python library that is high level runs as the main part of the Tensor Flow *** is used in the proposed work to implement deep learning *** comparative analysis of the proposed model and the existing work helps to show that there is a great improvement in the performance metrics in classifying the type of thyroid *** applying Adam and SGD(Stochastic Gradient Descent)optimizers in the training phase of the proposed model it was identified that these increase the operational efficiency of the modified ResNet *** retraining the model with SGD,the modified ResNet provides more accuracy of about 97%whereas the basic ResNet architecture attains 94%accuracy.A web-based frame-work is also developed which yields the type of thyroid disease as the output for a given input scanned image of the system.

关键词： Thyroid deep learning convolution neural network modified ResNet dual optimizer

来源：评论

学校读者我要写书评

暂无评论

Convolutional Neural Networks for MRI-Based Brain Tumor Classification 3

Convolutional Neural Networks for MRI-Based Brain Tumor Clas...

引用

3rd International Conference on Computation, Automation and Knowledge Management, ICCAKM 2022

作者： Soewu, Taiwo Singh, Dalwinder Rakhra, Manik Chakraborty, Gouri Shankar Singh, Arun Lovely Professional University Department of Computer Science and Engineering Phagwara India Lovely Professional University Department Computer Science and Engineering Phagwara India Lovely Professional University Assistant Professor Department of Computer Science and Engineering Phagwara India Lovely Professional University Phagwara Student IEEE Member Department of Computer Science and Engineering Punjab India

ISBN: (数字)9781665453196

ISBN: (纸本)9781665453196

Recently, the rate at which people die from cancer disease is very alarming. The major reason is because the cancer cells were not detected at the early stages. It may be difficult for medical professionals to evaluate the results of medical imaging techniques such as magnetic resonance imaging (MRI) because cancer cells may not be adequately represented for detection. This is why deep learning techniques are needed for earlier detection of cancer cells so as to reduce the death rate from *** this paper, we have applied the convolutional neural network on some MRI image dataset downloaded from Kaggle. The model was trained to be able to detect tumors in brain MRI images. The performance of the model was evaluated and it achieved 97.8% score for the accuracy, 98.5% specificity, 96.2% recall, 98.5% F1-score and 97.3% precision. A graph was drawn to compare the training loss to the validation loss, as well as the training accuracy to the validation accuracy. © 2022 IEEE.

关键词： Magnetic resonance imaging

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：