检索结果-内蒙古大学图书馆

Artificial Intelligence Based Security Improvement in Medical-IoT Health Care System Using Generative Deep Belief Neural Network

引用

SN computer science 2024年第8期5卷 1027页

作者： Anupriya, J. Devi, R. Renuga Department of Computer Science and Applications Faculty of Science and Humanities SRM Institute of Science and Technology Ramapuram Campus Tamil Nadu Chennai India

Artificial intelligence (AI) technology is the greatest option for medical applications since it improves data security and reliability. As a result, several AI-driven security procedures are implemented by standard IoT cloud framework initiatives. Medical Internet of Things (MIoT) devices have a role to play in meeting the urgent need for timely medical services, disease treatment, and monitoring of medical resources. Previous studies have demonstrated that health issues cannot be detected with sensitive features and cannot avoid data leakage, providing low security for sensitive data from medical IoT devices. So, to overcome the problems, an AI-based Generative Deep Belief Neural Network (AI -GDBN²) is to avoid the leakage of sensitive data and improve accuracy. We collect the dataset based on Medical-IoT features from the standard repository in the cloud;the initial step is preprocessing the data using normalization and cleaning, identifying missing values, eliminating inappropriate values, and reducing unscaled feature range weights that can provide the original dataset. The next step involves feature selection using the Fuzzy Recurrent Neural Network (FRNN) algorithm. This process utilizes the behavioral support factors of the features to determine their weights. Finally, the proposed AI-GDBN² model selects the attributes based on their maximum feature weight. AI provides security for sensitive data and avoids data leakages. And the AI-GDBN² is providing security based on the encryption method Advanced Honey Encryption Standard (AHES) to encrypt data using the maximum threshold values features and transmit paths. Next, using the randomly generated hash key, the AHES method encrypts the data. AI-GDBN² classifies the results based on sensitive and non-sensitive feature values and avoids abnormal data leakages to improve security. So, the simulation result proposed method AI-GDBN² shows an accuracy of 0.98%;th

关键词： Accuracy Artificial intelligence Data transmission Fitness function Machine learning Medical services Neural network Patient Sensor

来源：评论

学校读者我要写书评

暂无评论

Image Segmentation for MRI Brain Tumor Detection Using Advance AI algorithm

Image Segmentation for MRI Brain Tumor Detection Using Advan...

引用

2024 OPJU International Technology Conference on Smart Computing for Innovation and Advancement in Industry 4.0, OTCON 2024

作者： Karthikeyan, S. Lakshmanan, S. Hindustan Institute of Technology & Science Department. of Computer Applications Tamilnadu Chennai India

ISBN: (纸本)9798350373783

The existence of a brain tumor indicates an unusual growth of cells in the brain, manifesting as either benign (non-cancerous) or malignant (cancerous). Artificial Intelligence (AI) plays a crucial role in detecting and diagnosing brain tumors, primarily utilizing medical imaging techniques such as Magnetic Resonance Imaging (MRI). This initiative aims to employ Roboflow for data annotation and model training, utilizing the YOLOv8 algorithm to precisely handle the detection of brain tumors in MRI scans through image segmentation. The dataset utilized in this project encompasses annotated MRI scans illustrating tumor regions, meticulously prepared and enhanced through the application of Roboflow. YOLOv8, renowned for its expeditious and accurate object detection capabilities, is implemented and fine-tuned utilizing the annotated dataset. Model evaluation incorporates metrics such as Mean Average Precision (MAP), Recall, and Precision. Although YOLOv8 is inherently designed for object detection, this study explores adaptations for image segmentation, treating tumors as distinct classes. The iterative refinement process encompasses fine-tuning and optimization to elevate model performance and ensure robust generalization to novel data. The focus is on the advancement of MRI-based brain tumor detection through the utilization of cutting-edge deep learning techniques. This study highlights the potential of customizing YOLOv8 for the precise task of detecting brain tumors in MRI scans, establishing a foundation for future progress in the field of medical imaging. © 2024 IEEE.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Securing Industry 4.0: A Survey of Biometric Traits and Challenges 5

Securing Industry 4.0: A Survey of Biometric Traits and Chal...

引用

5th IEEE International Conference on Advances in Computing, Communication Control and Networking, ICAC3N 2023

作者： Kaur, Prabhjot Kant, Chander Kurukshetra University Department Of Computer Science & Applications Kurukshetra India

ISBN: (纸本)9798350330861

Industry 4.0, the advent of electronic industrial framework, is commonly referred to as the 4th industrial revolution and is progressing at highly growing rate as compared to the three former revolutions. Accessibility forms the fundamental basis for operation when there is a continuous exchange of knowledge between autonomous and efficient robots and business units. Workers play a vital role when utilizing emerging technology and exercising autonomy in decision-making. Employing techniques that separate individuals and their actions and reactions may be required. Enhancing safety measures, limiting access to restricted areas exclusively to approved personnel, and prioritizing employee welfare can yield advantages across the entire production chain. Hence, the collection of biometric data from individuals, whether with their awareness or without it, has the potential to facilitate authentication or monitor their emotions and cognitive states during their routine work-related tasks. After thoroughly examining the data, we have pinpointed three key domains where Industry 4.0 principles are integrated with the characteristics of biometric structures, traits and techniques. This review presents a inclusive overview of the many biometric functions used in the context of Industry 4.0, including their advantages, disadvantages, and practical uses. The study is now exploring new pathways to find innovative solutions. © 2023 IEEE.

关键词： Industry 4.0

来源：评论

学校读者我要写书评

暂无评论

SLDERCLC: Improving Smart Learning Frameworks Through Deep Learning Enhanced Recommender Systems Using an Efficient Cross-Layer Collaborative Approach

引用

SN computer science 2025年第2期6卷 1-18页

作者： Kaur, Harjinder Kaur, Tarandeep Kumar, Mukesh Bhardwaj, Vivek School of Computer Applications Lovely Professional University Punjab Phagwara 144411 India Department of Computer Applications Chandigarh School of Business Chandigarh Group of Colleges Jhanjeri Punjab Mohali 140307 India School of Computer Science and Engineering Manipal University Jaipur Jaipur 303007 India

The exponentially scaling domain of a smart learning frameworks necessitates advanced recommender systems capable of providing precise and relevant educational content to learners. Traditional recommender systems often grapple with challenges such as data sparsity, cold start problems, and the inability to capture complex user-item interactions & scenarios. This work introduces a novel approach to recommender systems in smart learning frameworks, addressing these limitations by leveraging deep learning techniques. Our proposed model amalgamates Deep Q Learning with Multilayer Graph Neural Networks (GNNs) for Cross Layer Collaborative Filtering and integrates Autoencoders with Capsule Networks for enhanced recommendation accuracy levels. The use of Deep Q Learning facilitates efficient decision-making in dynamic environments, while Multilayer GNNs adeptly handle relational data, improving the recommendation process by capturing the intricate connections within the educational contents. Furthermore, the incorporation of Autoencoders with Capsule Networks offers a sophisticated mechanism for understanding hierarchical relationships in data, which is crucial for personalized learning paths. The effectiveness of our model is substantiated through rigorous testing on the ASSISTments and EdNet datasets. The results are compelling, showcasing a 4.9% increase in precision, 5.5% improvement in accuracy, 3.5% higher recall, 2.9% greater AUC (area under the curve), 3.4% increased specificity, and a notable 4.5% reduction in delay for smart learning recommendations in comparison with SBBR, DRL and ROME. These improvements highlight the model’s proficiency in delivering timely and relevant educational content, thereby enhancing the learning experience of students as well as the teaching strategies of faculty. The framework attains significant improvements in the efficiency and effectiveness of educational content recommendations thereby increasing the retention rate of students w

关键词： Autoencoders Deep learning Graph Neural Networks Recommender systems Smart learning

来源：评论

学校读者我要写书评

暂无评论

Dynamic Resource Allocation Framework in Cloud Computing 5th

Dynamic Resource Allocation Framework in Cloud Computing

引用

5th International Conference on Soft Computing and Signal Processing, ICSCSP 2022

作者： Kaur, Gagandeep Chawla, Sonal Department of Computer Science and Applications Panjab University Chandigarh India

ISBN: (纸本)9789811986680

In Cloud Computing environment, resource provisioning, resource allocation, and resource scheduling are the most complex issues in these days. Cloud user expects best resource utilization and Cloud provider expects revenue maximization by considering budget and time constraints. In this research paper, Dynamic Resource Allocation algorithm has been proposed to allocate the resources to Cloud users and Cloud providers in dynamic Cloud environment. The proposed algorithm has been simulated and tested using CloudSim simulator tool. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Hybrid Classification Model for Network Traffic Density Estimation 4

Hybrid Classification Model for Network Traffic Density Esti...

引用

4th IEEE International Conference on Computing, Communication, and Intelligent Systems, ICCCIS 2023

作者： Mamta Poriye, Monika Upadhyaya, Shuchita Kurukshetra University Department of Computer Science and Applications Kurukshetra India

ISBN: (纸本)9798350306118

Traffic congestion is an increasing challenge in almost all parts of the globe. In the US alone, over $300B is incurred due to overcrowding. Existing work in the field of 'Network Traffic Density Estimation' is hampered by a lack of efficient methods for estimating network traffic density, as conventional methods have difficulty keeping up with the ever-changing topologies of today's networks. Optimizing network performance, boosting security, and allowing efficient resource allocation in complex network infrastructures all depend on finding a solution to this problem. The main cause of congestion at the network level is the increase in demand beyond the potential of the transport network. The machine learning techniques is proposed for the traffic density estimation. The machine learning technique is executed in 3 stages, such as to pre-process the data, extract the features and classify the data. This work suggests a hybrid model which is the combination of logistic regression and K-Nearest Neighbors. The proposed method achieved less error rate as compared to other ML techniques. The proposed method achieved 96.55% accuracy, 0.2% execution time and 3.45% error rate. Python is executed to visualize the suggested model with different metrics like accuracy, execution time, are considered to analyze the results. © 2023 IEEE.

关键词： Traffic congestion

来源：评论

学校读者我要写书评

暂无评论

Prediction of Thyroid Disease Using Machine Learning Algorithms 3

Prediction of Thyroid Disease Using Machine Learning Algorit...

引用

3rd International Conference on Advance Computing and Innovative Technologies in Engineering, ICACITE 2023

作者： Vinodhini, V. France, K. Hindustan Institute of Technology and Science Department of Computer Applications India

ISBN: (纸本)9798350399264

The detection and prognosis of diseases is one of the most significant uses of machine learning. Recently, disease classification and identification have benefited greatly and been significantly influenced by machine learning algorithms. One of these problems should cause people to worry about their health because the thyroid gland regulates metabolism, which is vital for maintaining human health. For the purpose of diagnosing thyroid disease, this study employed four machine learning methods: SVM, LR, DT & NB. According to this study, there are three types of thyroid disease: hyper, hypo, and normal. The laboratories of an Iraqi hospital provided the 2690 records that served as the basis for this investigation's data. Machine learning algorithms have shown promise in the diagnosis of thyroid disorders, assisting physicians and healthcare professionals in making an early diagnosis and improving the likelihood of therapy. Using all the features, Decision tree algorithm produced the greatest accuracy of 98.93%, however the Logistic regression approach earned the highest accuracy of 95.73% by removing three qualities (query thyroxine, query hypothyroid, and query hyperthyroid). © 2023 IEEE.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

An Enhanced Career Prospect Prediction System for Non-computer Stream Students in Software Companies 1st

An Enhanced Career Prospect Prediction System for Non-comput...

引用

1st International Conference on Computational Intelligence for Engineering and Management applications, CIEMA 2022

作者： Abraham, Biku Ambili, P.S. Department of Computer Applications Saintgits College of Engineering Kottayam India School of Computer Science and Applications REVA University Bangalore India

ISBN: (纸本)9789811984921

The stakeholders of the present education system have a key focus on the chances of on-campus employability of their wards. Since software companies play prominent role as campus recruiters, there is a low demand for the traditional Engineering branches. The need to enhance skills for software related employment is a challenge to the private sector professional institutions for the existence of non-computer streams in this scenario. In this study, the authors being academic experts in the similar field, attempt to support institutions, students, curriculum designers, and the faculty by providing insight to better employment possibility in campus. Data set for the study is from the same institution for which the placement prediction was completed. The methodology models the problem as a sequential event prediction problem and employs deep learning techniques. The system extracts data from data set with 19 attributes. The results indicate that this predictive approach provides enhancement methods by which a non-computer stream student can ensure full-time employment in software related jobs through in-campus selection. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for Plant Species Classification

Deep Learning for Plant Species Classification

引用

2023 International Conference on Sustainable Emerging Innovations in Engineering and Technology, ICSEIET 2023

作者： Kumar, M. Sunil Neeraja, Peddinti Gunamaneni, Anusha Nirmala, Ankireddy Vijaya Nikhil, Gobburu Department of Computer Science and Engineering AP Tirupati India Department of Computer Applications AP Tirupati India Sree Vidyanikethan Engineering College Department of Computer Applications AP Tirupati India

ISBN: (纸本)9798350329186

Plant species classification is a fundamental task in botany, agriculture, and environmental science, providing valuable insights into biodiversity, ecological systems, and agricultural practices. Botanists and the general public could benefit from an automated system for quickly recognizing plant species. The protection of biodiversity necessitates a better understanding of plant species identification [1]. Traditional methods of plant species identification relying on hand-crafted features can be complex and challenging, especially for non-specialists who have trouble remembering specific botanical words. A potential method has emerged for automatic plant species identification utilizing machine learning and deep learning techniques. Deep learning is effective in extracting features because it excels at giving more detailed information about images. This paper presents the implementation of Convolutional Neural Networks (CNNs), a deep learning-based approach that extracts features from leaf images and facilitates the classification of plant species. Deep learning methods, particularly CNNs, have demonstrated superior performance compared to handcrafted feature-based approaches. The ability of CNNs to automatically learn discriminative features from images has greatly improved the accuracy and efficiency of plant species classification. By leveraging deep learning methods, the proposed approach extracts features directly from leaf images, eliminating the need for manual feature engineering. The CNN architecture captures intricate patterns and structures present in the leaf images, enabling robust classification of plant species. The application of CNNs in plant species classification has yielded significant advancements in the field, providing a reliable tool for both experts and non-experts in identifying plant species. The implementation of CNNs facilitates the extraction of features from leaf images, leading to improved accuracy and efficiency. © 2023 IEEE.

关键词： Biodiversity

来源：评论

学校读者我要写书评

暂无评论

Customizing an Inference System to Extract Knowledge from Data Repositories Using Concepts of Fuzzy 12

Customizing an Inference System to Extract Knowledge from Da...

引用

12th International Conference on Current Trends in Advanced Computing, ICCTAC 2024

作者： Rathinasabapathy, R. Prathap, G. Madurai Kamaraj University Department of Computer Applications Tamilnadu Madurai India Kristu Jayanti College Department of Computer Science Karnataka Bangalore India

ISBN: (纸本)9798350395808

Algorithms are in abundance for extraction of information but some are not up to the expectation. It is due to the lack of effective mechanism in place to represent the required information. Fuzzy logic provides a sophisticated way of expressing the requirement with its membership functions, rule base and reasoning mechanisms. Membership function helps to flexibly specify the grade and extent of information needed and the rule base provides methods for combining them for effective representation of what is required. The reasoning mechanism in fuzzy logic helps to convert these representations into appropriate values so that the information can be classified according to our needs. In this research endeavour, a methodology has been delineated to describe the required information and a fuzzy model has been defined to extract the documents. It is flexible to change the required information and its extent using fuzzy logic. Using different types of rule sets how the extraction of information is being changed, has been discussed. Fuzzy inference system has been modelled to rank the documents in the order of the grade of information content. An algorithm has been provided to extract the information from rule sets. A generalized membership function has been delineated in this research work. By changing the parameter values in the membership function different membership functions can be initiated. By using two different sets of parameter values, two different membership functions have been modelled in this research work. These values combined with two different inference models namely, mamdani and sugeno models, totally four different fuzzy models have been implemented in this research work and the results are shown in the graphs at the end of this research work. © 2024 IEEE.

关键词： Fuzzy control

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：