检索结果-内蒙古大学图书馆

Enhancing IoT gateway management security: a deep learning based framework for intrusion detection and threat mitigation

引用

International Journal of Information Technology (Singapore) 2025年 1-11页

作者： Sonal Deswal, Suman Department of Computer Science and Engineering Deenbandhu Chhotu Ram University of Science and Technology Murthal Haryana Sonipat 131039 India

The proposed scheme in the paper focuses on the stacked hybrid deep learning model, which includes LSTM (long-short term memory), GRU (gated recurrent unit), SimpleRNN (recurrent neural network), and CNN (convolutional neural network) architectures for enhancing intrusion detection in IoT (Internet of Things) gateway management systems. This work leverages these neural networks in a union to learn both temporal and spatial features from network traffic that enable the model actually to detect complex cyber threats. The model produced showed good performance, with 99.9% in training accuracy, 99.84% in testing, and effective precision, recall, and F1-score. Its robustness was further evaluated against some of the state-of-the-art approaches. Noteworthy improvements included all major evaluation metrics like MCC (Matthews correlation coefficient), MSE (Mean square error), and AUC (Area under curve). Moreover, comparative analysis of the proposed model has been done with latest state of art work on the basis of metrics like precision, recall, F1 score and accuracy. The results confirm the model has a high potential for improving security in both IoT and WSN (wireless sensor network) environments. Therefore, it will be very effective in real-world intrusion detection. © Bharati Vidyapeeth's Institute of computer Applications and Management 2025.

关键词： Cyber threats Deep learning IoT Network traffic WSN

来源：评论

学校读者我要写书评

暂无评论

Transformers-Based Deep Learning for Parkinson's Disease Detection Using Electroencephalography and Stockwell Transform 3

Transformers-Based Deep Learning for Parkinson's Disease Det...

引用

3rd International Conference on Machine Learning and Data engineering, ICMLDE 2024

作者： Afonso, Melina Maria Edla, Damodar Reddy Ramesh, Dharavath Department of Computer Science and Engineering National Institute of Technology Goa Goa Cuncolim403703 India Department of Computer Engineering Goa College of Engineering Goa Ponda403401 India Department of Computer Science and Engineering IIT Dhanbad Jharkhand Dhanbad826004 India

Neurodegenerative conditions such as Parkinson's disease (PD) have, in recent times, seen a major boost to their understanding due to the use of deep learning (DL) in the analysis and diagnosis process. DL, in particular, has been effective in unearthing many complex and intricate concepts of the EEG signal. This research is an attempt to understand the impact of the state-of-the-art DL technology, the transformer, in the diagnosis of PD from electroencephalography (EEG) signals. The time-domain EEG signals are transformed into time-frequency domain using the Stockwell transform for better insights. This transformation produces an image dataset, which can now be analyzed intricately for time and frequency features. These images are then fed to three transformer models tailored specifically for PD detection based on the response of the model. The classical vision transformer (ViT), the compact convolution transformer (CCT), and the swin transformer are selected for the purpose. The hyperparameters of each of the models are tuned by evaluating the various performance parameters at every iteration to design the most efficient models. The performance of the models is compared to each other as well as to other recent research using similar technology. The publicly available San Diego dataset is used for the analysis, and the ViT displays a commendable accuracy of 94% with the loss closing in towards zero. The CCT also performs well with fewer computational requirements. © 2024 The Authors. Published by ELSEVIER B.V.

关键词： Neurodegenerative diseases

来源：评论

学校读者我要写书评

暂无评论

Mental Health Monitoring Using Deep Learning Technique for Early-Stage Depression Detection

引用

SN computer science 2023年第6期4卷 701页

作者： Singh, Khushboo Ahirwal, Mitul Kumar Pandey, Manish Department of Computer Science and Engineering MANIT Bhopal India

An electroencephalogram, often known as an EEG, can detect neuronal activity by analysing the electrical currents that are generated within the brain by a collection of specific pyramidal cells as a result of the synchronised activity. EEG signal contains essential information about brain activity and is often used to diagnose and treat brain diseases such as depression and other healthcare issues. This study offers a deep learning model that is built on convolutional neural network (CNN) and bi-directional long short-term memory (Bi-LSTM), with the goal of extracting negative emotions from EEG data, such as depression. The depression recognition deep learning model (DRDL) includes segmentation, normalization, data augmentation and ensembling methods to preprocess and enhance the number of samples in EEG signals. The DRDL model employed 1D-CNN and Bi-LSTM to classify human EEG signals as Relaxed, Mild depression, Moderate depression, and Severe depression. The DRDL model achieves 91.31% and 90.12% accuracy, respectively, over binary and quaternary class labeling. Furthermore, the classification result significantly improves the performance of lower-performing subjects 1 and 23 in the DEAP dataset. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Bi-LSTM Classification Convolutional neural network Deep learning Depression Electroencephalogram Emotion Healthcare

来源：评论

学校读者我要写书评

暂无评论

PneumonXNet - A deep learning model for pneumonia detection

PneumonXNet - A deep learning model for pneumonia detection

引用

2024 IEEE International Conference on Signal Processing and Advance Research in Computing, SPARC 2024

作者： Aditya, Harshit Ramani, Anchita Kumar, Nilesh Kasturi Bhattacharjee, Vandana Sikkim Manipal Institute of Technology Department of Computer Science and Engineering Majitar India Birla Institute of Technology Department of Computer Science and Engineering Mesra Ranchi India

ISBN: (纸本)9798350385199

Standard chest X-rays are used to diagnose pneumonia, an infection of the lungs that can be highly contagious due to being caused by viruses and bacteria. Nevertheless, this method is more time-consuming and routine-dependent with a bias in outcome. So to overcome these problem we developed PneumonXNet, a bespoke CNN model for the detection of pneumonia. We evaluated the performance of our method with respect to pre-trained models such as VGG16, EfficientNetB3 and ResNet50 on a X-ray image dataset comprising 5,856 images. The best result we have reached is a test accuracy of 97% with dropout value of 0.4 which is better than the pretrained models. The pre-trained model EfficientNetB3 showed the highest accuracy, which points on possible applications of transfer learning. This research signifies the capabilities of CNN models to streamline and improve detection on pneumonia, which in turn paves way for technological sophistication in health care sectors. The findings of this study carry significant impact on pneumonia detection, emphasizing the importance of deep learning model in pneumonia detection and how it can help in saving a lot of time for doctors and patients from the traditional X rays collecting systems. This can also reduce the overall treatment cost for patients. This research therefore offers new avenues for developing strong techniques for pneumonia identification, which hold promise for improving healthcare technology and patient outcomes in general. © 2024 IEEE.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Smart Nutrient Deficiency Prediction System for Groundnut Leaf

引用

Intelligent Automation & Soft Computing 2023年第5期36卷 1845-1862页

作者： Janani Malaisamy JebakumarRethnaraj Department of Computer Science and Engineering School of ComputingSRM Institute of Science and TechnologyChengalpattuTamil Nadu603203India

Prediction of the nutrient deﬁciency range and control of it through application of an appropriate amount of fertiliser at all growth stages is critical to achieving a qualitative and quantitative *** fertiliser in optimum amounts will protect the environment’s condition and human health *** identiﬁcation also prevents the disease’s occurrence in groundnut crops.A convo-lutional neural network is a computer vision algorithm that can be replaced in the place of human experts and laboratory methods to predict groundnut crop nitro-gen nutrient deﬁciency through image *** chlorophyll and nitrogen are proportionate to one another,the Smart Nutrient Deﬁciency Prediction System(SNDP)is proposed to detect and categorise the chlorophyll concentration range via which nitrogen concentration can be *** model’sﬁrst part is to per-form preprocessing using Groundnut Leaf Image Preprocessing(GLIP).Then,in the second part,feature extraction using a convolution process with Non-negative ReLU(CNNR)is done,and then,in the third part,the extracted features areﬂat-tened and given to the dense layer(DL)***,the Maximum Margin clas-siﬁer(MMC)is deployed and takes the input from DL for the classiﬁcation process toﬁnd *** dataset used in this work has no visible symptoms of a deﬁciency with three categories:low level(LL),beginning stage of low level(BSLL),and appropriate level(AL).This model could help to predict nitrogen deﬁciency before perceivable *** performance of the implemented model is analysed and compared with ImageNet pre-trained *** result shows that the CNNR-MMC model obtained the highest training and validation accuracy of 99%and 95%,respectively,compared to existing pre-trained models.

关键词： CNN chlorophyll groundnut crop deep learning nitrogen deﬁciency

来源：评论

学校读者我要写书评

暂无评论

Enhancing intrusion detection in IIoT: optimized CNN model with multi-class SMOTE balancing

引用

Neural Computing and Applications 2024年第24期36卷 14643-14659页

作者： Eid, Abdulrahman Mahmoud Soudan, Bassel Nassif, Ali Bou Injadat, MohammadNoor Department of Computer Engineering College of Computing and Informatics University of Sharjah Sharjah United Arab Emirates Department of Data Science and AI Faculty of Information Technology Zarqa University Zarqa Jordan

This work introduces an intrusion detection system (IDS) tailored for industrial internet of things (IIoT) environments based on an optimized convolutional neural network (CNN) model. The model is trained on a dataset that was balanced using a novel multi-class implementation of synthetic minority over-sampling technique (SMOTE) that ensures equal representation of all classes. Additionally, systematic optimization will be used to fine tune the hyperparameters of the CNN model and mitigate the effects of the increased size of the training dataset. Evaluation results will demonstrate substantial improvement in performance when the optimized CNN model is trained on the balanced dataset. The proposed IDS will be evaluated using the IIoT-specific WUSTL-IIOT-2021 dataset, and then its generalization capability will be verified using the non-domain specific UNSW_NB15 dataset. The model’s performance will be evaluated using accuracy, precision, recall, and F1-score metrics. The results will demonstrate that the proposed IDS is highly effective with performance exceeding 99.9% on all performance metrics. The IDS is also highly effective in detecting intrusion for generic IT networks achieving improvements in excess of 30% compared to the default baseline model. The results emphasize the versatility and effectiveness of the proposed IDS model, making it a reliable and adaptable solution for enhancing network security across diverse network environments. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

Improved clustering techniques for paediatric cerebral palsy gait assessment during rehabilitation

引用

International Journal of Information Technology (Singapore) 2024年 1-18页

作者： Singhal, Prateek Yadav, Rakesh Kumar Department of Computer Science and Engineering Maharishi University of Information Technology Lucknow India

The gait abnormality may be the cause of various diseases like foot drop, lower back trembling, and osteoarthritis in the human body. The causes may affect body performance. The problem may be solved if we notice it between the ages of 2 and 20. Today's medical research struggles to identify normality and abnormality in children at a young age. The gait abnormalities depend on a complex neurological condition called cerebral palsy. The article proposes an improved fuzzy C-mean-PSO technique and also defines selection criteria for gait patterns, such as optimal number identification gait profiles, mean square error, silhouette coefficient, and Dunn index. The researcher used 156 patients’ data from the available O’Malley gait dataset for experimental purposes. We partitioned 156 patients into 5 different combinations. In the first two combinations, we applied conventional methods, and the next three employed proposed methods. Finally, we found the 91.6% CPI (Cluster Purity Index), that is greater than existing techniques. In the future, we can perform the proposed methods on various datasets. The findings indicate that employing clustering-based gait profiles improved fuzzy C-mean-PSO optimised using these methods can aid in measuring clinical rehabilitation for children with cerebral palsy. © Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： Gait analysis Gauss Cuckoo search IPSOFCM PSO Rehabilitation

来源：评论

学校读者我要写书评

暂无评论

Estimating the distance to an object from static color stereo images using convolutional neural networks 27

Estimating the distance to an object from static color stere...

引用

27th International Conference on Knowledge Based and Intelligent Information and engineering Sytems, KES 2023

作者： Kulawik, Joanna Faculty of Mechanical Engineering and Computer Science Department of Computer Science Czestochowa University of Technology Dabrowskiego 69 Czestochowa42-201 Poland

In issues such as autonomy of movement, the vital information is the vehicle's distance from the obstacle. The research presented is devoted to the phenomenon of disparity in color stereo images. Their goal was to estimate the distance to the obstacle. A convolutional neural network regression model was used as a research method. The data set is a proprietary set of color stereo static images. Testing the learned CNN model gave excellent results. The analysis of the obtained results showed that the developed solution could be used in engineering solutions such as supporting vehicle control. © 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://***/licenses/by-nc-nd/4.0)

关键词： artificial intelligence deep learning estimating distance stereo-vision

来源：评论

学校读者我要写书评

暂无评论

Prediction in Multivariate Time Series Data using Generative Adversarial Network 15

Prediction in Multivariate Time Series Data using Generative...

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Medhi, Shrabani Kashyap, Samudra Das, Nitul Girijananda Chowdhury University Department of Computer Science and Engineering Guwahati India Girijananda Chowdhury Institute of Management and Technology Department of Computer Science and Engineering Guwahati India

ISBN: (纸本)9798331300579

Air pollution is an issue of great concern. PM2.5 is the most dangerous pollutant out of all the pollutants. A large number of missing values is present in multivariate pollution data. This makes the prediction of PM2.5 concentration very difficult. Traditional approaches to deal with missing values includes mean imputation, median imputation, case deletion, matrix factorization-based imputation, case deletion, etc. All these methods are not capable of modelling the temporal dependencies and complex distributed nature of multivariate pollution data. We have dealt with missing values by mean imputation, median imputation, Generative Adversarial Network imputation and Conditional Tabular Generative Adversarial Network. We have used Artificial Neural Network to predict PM2.5 concentration. The results reveal that for time series data, imputation by median model has performed waste. No significance improvement is seen in prediction of PM2.5 concentration using GAN and CTGAN model if we compare it with imputation by mean model. For time series data, more research work needs to be done with GAN and its variants so that better results can be achieved. © Grenze Scientific Society, 2024.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Telugu Language Analysis with XLM-RoBERTa: Enhancing Parts of Speech Tagging for Effective Natural Language Processing

引用

SN computer science 2025年第2期6卷 1-13页

作者： Bharathi Mohan, G. Prasanna Kumar, R. Krishna Jayanth, K. Doss, Srinath Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Chennai India Faculty of Engineering and Technology Botho University Gaborone Botswana

This research investigates the efficacy of XLM-RoBERTa, a potent deep learning architecture rooted in transformer networks, for Part-of-Speech (POS) tagging—a foundational task in Natural Language Processing (NLP). The study narrows its focus to address the formidable UCD Telugu dataset, renowned for its intricate morphological characteristics that pose challenges to accurate tagging. Through meticulous fine-tuning of the pre-trained XLM-RoBERTa model. This study adeptly discerns and exploit intricate syntactic and semantic patterns underpinning the interplay between lexical elements and their respective POS tags. The rigorous evaluation regimen, complemented by a comprehensive statistical comparison with the Multilingual BERT model, conclusively establishes XLM-RoBERTa’s superiority. This advanced model attains a remarkable accuracy score of 91%, decisively surpassing the Multilingual BERT model, which achieves an accuracy of 88%. These findings not only underscore XLM-RoBERTa’s prowess in precise POS tagging, especially in the context of morphologically rich languages like Telugu, but also carry significant implications for the broader NLP community. This research unlocks valuable insights and positions XLM-RoBERTa as an exemplar of reliability and robustness in POS tagging. Furthermore, it extends an invitation for further exploration and innovation in deep learning architectures within the NLP domain. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Bert Multilingual Bert NLP Parts-of-speech tagging Telugu XLM-RoBERTa

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：