检索结果-内蒙古大学图书馆

4th International Conference on Technological Advancements in Computational sciences, ICTACS 2024

作者： Soni, Tanishq Chitkara University Institute of Engineering and Technology Chitkara University Department of Computer Science and Engineering Punjab India

ISBN: (纸本)9798350387490

Strawberry leaf diseases seriously affect crop output and quality;hence quick and precise identification techniques are quite important. This work presents a deep learning-based method based on LeNet convolutional neural network (CNN) architecture to detect strawberry leaf illnesses. Originally meant for handwritten digit classification, the LeNet model has been modified and refined to properly identify illnesses in strawberry leaves. High-resolution photos of strawberry leaves, categorized as healthy and diseased, make up the dataset utilized for training and validation. The LeNet model shows a remarkable accuracy of 95% on the validation set, therefore showing its potential as a dependable tool for the diagnosis of strawberry leaf disease from experimental results. Key measures including precision, recall, Fl-score, and a confusion matrix were used to assess the model's performance, thereby verifying its dependability and efficiency in separating healthy from sick leaves. © 2024 IEEE.

关键词： CNN Detection Leaf Disease LeNet Strawberry

来源：评论

学校读者我要写书评

暂无评论

Analysing the Accuracy of Detecting Phishing Websites using Ensemble Methods in Machine Learning 3

Analysing the Accuracy of Detecting Phishing Websites using ...

引用

3rd International Conference on Artificial Intelligence and Smart Energy, ICAIS 2023

作者： Menaka, S. Harshika, Jonnalagadda Philip, Sarah John, Rashi Bharathiraja, N. Murugesan, S. Srm Institute of Science and Technology Department of Computer Science and Engineering Chennai India Chitkara University Institute of Engineering and Technology Department of Computer Science and Engineering Chennai India Tagore Engineering College Department of Information Technology Rathinamangalam Chennai India

ISBN: (纸本)9781665462167

Phishing attacks are now one of the prevalent dangers that firms, service providers and internet users must deal with. Rather than targeting software vulnerabilities, it targets human vulnerabilities. It is the act of enticing users to attain their personal data using fake emails and websites. Like how e-commerce sectors are growing, phishing attacks are also developing. Preventing phishing attempts is a critical aspect of protecting online transactions. Since hacktivists, spy agencies and cybercriminals now have a rich field in which they can operate sophisticated phishing attacks, prompt detection of phishing attempts is more critical than ever. To properly respond to various phishing attacks, it is required to gain a thorough understanding of these attacks, and suitable response techniques must be used. The challenges faced in this research is finding the appropriate datasets and Feature extraction prompted the study of several modules, in addition to understanding every module and attaining the desired outcome from it. Machine learning techniques are used to accurately identify phishing attacks before cause harm to a user. Being able to handle the changing nature of phishing attempts and offering an accurate method of classification, it is one of the most practical ways to approach the situation. © 2023 IEEE.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

Enhancing Brain Stroke Risk Prediction with Multi-Algorithm Evaluation and Web Interface

Enhancing Brain Stroke Risk Prediction with Multi-Algorithm ...

引用

2024 International Conference on engineering and Computing, ICECT 2024

作者： Ahmed, Muhammad Liaquat, Umer Rehman, Attique Ur Javaid, Sabeen Ali, Tahir Muhammad Mir, Azka University of Sialkot Department of Software Engineering Sialkot Pakistan Gulf University for Science and Technology Department of Computer Science Kuwait

ISBN: (纸本)9798350349719

Brain stroke is the world's leading cause of death, impacting numerous lives annually. The chances of having a stroke have increased by 50% over one's lifetime, impacting one in four people worldwide. Machine learning may assist in the prior detection of strokes, allowing for more timely diagnosis which can save a person's life and disability. However, existing methods face challenges in integrating ML models with clinical applications to predict patient stroke status with optimal accuracy. This study conducts a comprehensive investigation into brain stroke risk prediction using numerous ML algorithms, including Random Forest, Logistic Regression, KNN, Naive Bayes, XGBoost, and LightGBM. Our main contribution is the development of a web app that predicts stroke based on user inputs with optimal accuracy. The selected dataset contains 5110 records of normal people and 249 records of people having strokes which is highly imbalanced. To ensure model robustness, the dataset undergoes balancing through SMOTE oversampling and KNN Imputation for handling missing values. Custom thresholds for classifiers optimize training, with comprehensive evaluations utilizing key statistics such as confusion matrices, F1-Score, recall, accuracy, and precision. Cross-validation is employed to assess potential overfitting and underfitting risks in machine learning models. Random Forest ranked as the most imposing, reaching a remarkable 99.85% accuracy on the test set with a custom threshold of 0.6 and a random state at 42. Additionally, a user-friendly Web App has been developed to predict stroke status based on user input, enhancing practical usability and accessibility for individuals seeking risk assessments. © 2024 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Model view controller (MVC) architecture and client-to-client file transfer protocol with binarized spiking neural network for building a smart home appliances control in IoT

引用

International Journal of Information technology (Singapore) 2023年第6期15卷 3189-3200页

作者： Yadawad, Ranganath Kulkarni, Umakant P. Department of Computer Science and Engineering SDM College of Engineering and Technology Karnataka Dharwad 580002 India

In the time Internet of Things (IoT) the term “Smart” is mostly used because of its application in various entities such as smart cities, smart homes, and smart cars. These are done by integrating Machine Learning (ML) algorithms along with IoT devices. In the context of the smart home, many researchers use IoT designs with Deep Learning (DL) techniques, although the majority of them are limited because of hardware issues of the end devices. So, the transport of detected messages from IoT devices to the application layer takes longer when object detection techniques are used in real-time systems, which lowers the DL’s accuracy. To overcome these issues a model view controller (MVC) architecture and client-to-client file transfer protocol (C2CFTP) with binarized spiking neural network (BS4NN) for building a smart home appliances control in IoT is proposed in this manuscript. In this smart home automation system, users can remotely monitor their houses by employing an object detection algorithm built on the client-to-client model view controller (C2CMVC) architecture. C2CMVC architecture consists of three layers such as bottom Layer, Repository layer, and application layer. This, IoT-based sensing device, collects data from a laptop or personal computer (PC) that has a camera attached using the bottom layer. For object detection, the BS4NN is employed, and when an object is discovered, a message is delivered to the C2CFTP. The middle layer, also known as the repository layer, in this, the C2CFTP broker used. C2CFTP is a novel protocol which is efficiently transfers messages to the application layer. Users receive notifications by short message service (SMS) and email, and the system's application layer sends a command message. The performance of the proposed method attains 32.0%, 1.0%, 15.0%, and 15.0% lower delay is likened to the existing methods. © 2023, The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of computer Applications and Management

关键词： Binarized Spiking Neural Network Client-to-client file transfer protocol Internet of Things Model view controller architecture Object detection

来源：评论

学校读者我要写书评

暂无评论

Attention-Guided CNN with LSTM for Temporal Gastrointestinal Disease Progression Prediction in Endoscopic Video Sequences 4

Attention-Guided CNN with LSTM for Temporal Gastrointestinal...

引用

4th IEEE International Conference on Mobile Networks and Wireless Communications, ICMNWC 2024

作者： Mohamad, Fayazoddin Dhumpati, Raghu Reenadevi, R. Rao, P.S.V. Srinivasa Sutaria, Kamal Kulkarni, Shirish NY United States Bahrain Polytechnic Department of Computer Science & Engineering Bahrain City Bahrain Salem India Joginpally B R Engineering College Department of Computer Science and Engineering Hyderabad India Parul institute of Engineering and Technology Parul University Faculty of Engineering and Technology Department of Computer Science and Engineering Post Limda India Faculty of Law Pune India

ISBN: (纸本)9798350352931

The detection and tracking of changes in the progression of GI disease using endoscopic video analysis remains difficult due to temporal changes and image complexity. Proper models of prediction are critical in diagnosing individuals at an early stage and offering them treatment. In the work presented in this paper, we introduce novel Attention-Guided Convolutional Neural Network (AG-CNN) combined with LSTM networks to analyze temporal endoscopic video sequences. It uses spatial attention to catch necessary areas and temporal attention to capture assertive illness transformation in video frames. For disease progression prediction the proposed model yield 94.8% accuracy with 93.2% sensitivity and 95.6% specificity. The attention mechanism increased the feature extraction speed by 22 percent when compared to traditional CNN-LSTM classes. Independent validation checks confirmed that the two models produced a prediction concordance of 92.7% based on different patient samples. The overall performance of temporal GI disease progression prediction by the AG-CNN-LSTM framework shows potential usage in clinical diagnosis by providing automated disease progression monitoring and treatment planning. © 2024 IEEE.

关键词： Video analysis

来源：评论

学校读者我要写书评

暂无评论

Appearance Based Dynamic Hand Gesture Recognition Using 3D Separable Convolutional Neural Network

引用

computers, Materials & Continua 2023年第7期76卷 1213-1247页

作者： Muhammad Rizwan Sana Ul Haq Noor Gul Muhammad Asif Syed Muslim Shah Tariqullah Jan Naveed Ahmad Department of Electronics University of PeshawarPeshawar25120Pakistan Department of Electronics Engineering Korea Polytechnic UniversitySiheungKorea Department of Electrical Engineering Capital University of Science and TechnologyIslamabad44000Pakistan Department of Electrical Engineering University of Engineering and TechnologyPeshawar25120Pakistan Department of Computer Science Prince Sultan UniversityRiyadh11586Saudi Arabia

Appearance-based dynamic Hand Gesture Recognition(HGR)remains a prominent area of research in Human-computer Interaction(HCI).Numerous environmental and computational constraints limit its real-time *** addition,the performance of a model decreases as the subject’s distance from the camera *** study proposes a 3D separable Convolutional Neural Network(CNN),considering the model’s computa-tional complexity and recognition *** 20BN-Jester dataset was used to train the model for six gesture *** achieving the best offline recognition accuracy of 94.39%,the model was deployed in real-time while considering the subject’s attention,the instant of performing a gesture,and the subject’s distance from the *** being discussed in numerous research articles,the distance factor remains unresolved in real-time deployment,which leads to degraded recognition *** the proposed approach,the distance calculation substantially improves the classification performance by reducing the impact of the subject’s distance from the ***,the capability of feature extraction,degree of relevance,and statistical significance of the proposed model against other state-of-the-art models were validated using t-distributed Stochastic Neighbor Embedding(t-SNE),Mathew’s Correlation Coefficient(MCC),and the McNemar test,*** observed that the proposed model exhibits state-of-the-art outcomes and a comparatively high significance level.

关键词： 3D separable CNN computational complexity hand gesture recognition human-computer interaction

来源：评论

学校读者我要写书评

暂无评论

K-Anonymization approach for privacy preservation using data perturbation techniques in data mining

引用

Materials Today: Proceedings 2022年 64卷 578-584页

作者： Kiran, Ajmeera Shirisha, N. Department of Computer Science and Engineering MLR Institute of Technology Hyderabad India

Nowadays, privacy is a big issue for everyone because the internet generates and shares massive amounts of data every day. Because traditional privacy-preserving techniques do not protect sensitive data well enough, malicious attacks on sensitive data are possible. Traditional methods may have many drawbacks due to various attacks on sensitive data. Thus, data must be preserved before being shared with others. K-Anonymity protects privacy. By definition, anonymization means that a person is not identifiable, traceable, or reachable. Anonymization using k-anonymity protects data privacy and protects entire datasets. The traditional data perturbation method only works with numerical data, but the proposed method is designed to work with categorical data. The Proposed Framework perturbs data for multiple columns at once. The proposed framework's performance is evaluated in two ways. Initially, it keeps the data mining model accurate. Second, it protects the original data privacy while minimizing data loss. The main motivation for using k-anonymity is to remove or transfer personally identifiable information. © 2022

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Spatial Analysis and Morphological Clustering of Barchan Dunes using Advanced Computational Techniques 3

Spatial Analysis and Morphological Clustering of Barchan Dun...

引用

3rd International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2025

作者： Shobha, P. Sujay, B.R. Bennur, Swarna Girish Nalini, N. Raghunandan, K.R. Department of Computer Science and Engineering Nitte Meenakshi Institute of Technology Benagaluru India Nitte India

ISBN: (纸本)9798331527549

Barchan dunes, crescent-shaped aeolian landforms shaped by unidirectional wind regimes, are integral to understanding sediment transport and landscape evolution in arid regions. This study conducts a geomorphometric analysis and clustering of barchan dunes, focusing on their spatial distribution, morphological characteristics, and environmental controls. Leveraging high-resolution digital elevation models (DEMs) and advanced clustering algorithms such as K-Means, DBSCAN, and Gaussian Mixture Models, the research identifies patterns in dune morphology and examines their relationships with local topography and climatic drivers. The methodology integrates preprocessing steps, including feature extraction and scaling, with unsupervised clustering to classify dunes based on key attributes such as width, length, height, and geographic location. Morphometric analyses reveal spatial and structural variations, while clustering results highlight patterns linked to wind regimes and sediment availability. Among the clustering techniques, DBSCAN emerges as the most effective for identifying irregularly shaped clusters, achieving higher evaluation metrics such as Silhouette Scores and Davies-Bouldin Index values. This study investigates the morphology and spatial distribution of barchan dunes using advanced clustering techniques such as DBSCAN and K-Means. The primary issue addressed is the challenge of accurately classifying and analyzing irregularly shaped dune formations under varying environmental conditions. Utilizing high-resolution digital elevation models (DEMs), the research identifies distinct morphological patterns and their correlation with wind regimes and sediment availability. The study finds that DBSCAN outperforms other clustering algorithms in handling the irregular geometries of dunes, achieving superior evaluation metrics like Silhouette Scores and Davies-Bouldin Index. These findings provide valuable insights for environmental monitoring and desertification m

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach to Save Energy by Detecting Faulty HVACs

引用

Journal of The Institution of Engineers (India): Series B 2022年第2期103卷 305-311页

作者： Agarwal, Anshul Department of Computer Science and Engineering National Institute of Technology Nagpur India

Studies have shown that heating ventilation and air-conditioning (HVAC) systems are the major contributors to the high energy consumption of buildings. This has some serious deleterious effects on the environment. Thus, to reduce and optimize the energy consumption, it is important for the HVACs to work efficiently at all times. Existing techniques have focused on the optimal usage of HVACs for minimizing the energy consumption of buildings. This paper offers a pragmatic and holistic solution for saving energy in smart buildings by detecting faulty HVACs using power consumption and temperature data only. The developed approach is applied to a building consisting of forty-eight HVACs and is situated in Mumbai. A total of thirty-one HVACs were detected to be faulty. The results indicate that about 46% of energy can be saved if all the HVACs operating in the building are free of faults. This is an impressive saving in energy consumption, which demonstrates the effectiveness of this novel approach. © 2021, The Institution of Engineers (India).

关键词： Energy conservation

来源：评论

学校读者我要写书评

暂无评论

Enhance Intrusion Detection by Analyzing the Behavior of Labeled and Unlabeled Classification to Obtain Better Accuracy 2

Enhance Intrusion Detection by Analyzing the Behavior of Lab...

引用

2nd International Conference on Advanced Computing and Communication Technologies, ICACCTech 2024

作者： Suriya Prakash, J. Srinidhi, N. Latha, A. Chaithra Kiran, S. Department of Computer Science and Engineering Karnataka Bengaluru India Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Department of Computer Science and Engineering Manipal India Sri Krishna Institute Of Technology Visvesvaraya Technological University Department of Computer Science and Engineering Karnataka Bengaluru India KSIT Department of Computer Science and Design Karnataka Bengaluru India Nitte Meenakshi Institute of Technology Department of Mathematics Karnataka Bangalore India

ISBN: (数字)9798331519056

ISBN: (纸本)9798331519056

Intrusion Detection System has accelerated globally as a result of the need to identify intrusions that occur in network data flow. The two IDS methods that are primarily used by machine learning and network pattern detection are anomaly-based and signature-based methods. A network can be made foolproof by keeping it safe by keeping an eye out for any hazardous activity. In network dataflow systems, network traffic can be produced via simulators. Simulators are a useful tool for detecting harmful behavior. Intrusions that are host-based, protocol-based, application protocol-based, network-based, and hybrid-based are some of the most prevalent kinds that have been discovered. In this paper,open source datasets ISP datasets and WIDE datasets are employed for experimentation. The crux of this paper is to carry out network classification by using minimal training sets and classifying maximal test sets for procuring accurate results. Lastly, the Labelled and Unlabelled technique is used to compare various accuracy outcomes. The suggested scheme's effectiveness is confirmed by the empirical investigation conducted on ISP and WIDE data. In this paper, the labeled and unlabeled classification proposed here identifies the intruder application with minimal labeled dataset. Most importantly automatic labeling of large test sets with small training set is accomplished. The analysis process for classification methods involving ISP and WIDE datasets by an experimental way has produced efficient classification results. The Labeled classification algorithm produces 93% accuracy and the unlabeled classification algorithm produces 84% accuracy. © 2024 IEEE.

关键词： Network intrusion

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：