检索结果-内蒙古大学图书馆

5th International Symposium on Signal and Image Processing, ISSIP 2024

作者： Singh, Jagendra Saraswat, Mala Upreti, Kamal Agrawal, Gaurav Hasan, Dler Salih Shaik, Nazeer School of Computer Science Engineering and Technology Bennett University Greater Noida India School of Computer Science Engineering & Technology Bennett University Greater Noida India Delhi-NCR Ghaziabad India KIET Group of Institution Ghaziabad India Department of Computer Science & Information Tech. University of Salahaddin-Erbil Erbil Iraq Department of Computer Science & Engineering Srinivasa Ramanujan Institute of Technology Autonomous Anantapur India

ISBN: (纸本)9789819795147

The main objective of this research was the development and evaluation of an IoT- and machine learning-based health monitoring system capable of protecting patient’s vital organs through cloud diagnosis. This could be achieved by connecting a set of sensors, including temperature, pressure, heart rate, and oxygen sensors, to the patient and allowing them to communicate with the cloud to transmit real-time data via IoT technologies. The data could be further analyzed and predicted using cloud-based machine learning algorithms. This study investigated the performance of different machine learning models, including Artificial Neural Networks (ANN), Support Vector Machines (SVM), Naive Bayes (NB), and Negative Decision Trees (DT), for the purpose of patient’s health prediction based on the sensor data. After experimentation and evaluation, we found that the ANN model demonstrated the best predictive ability, with an accuracy level of 99.45%. The SVM, NB, and DT models also demonstrated good performance, with the accuracies of 96.5%, 94.34%, and 91.2%, respectively. Therefore, this research demonstrated that IoT and machine learning technologies could be successfully employed in healthcare for remote patient monitoring and timely prediction. The created system allows for real-time monitoring, which enables early prediction, potentially leading to improved patient outcomes, cost savings, and higher efficiency of provided care. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Remote patient monitoring

来源：评论

学校读者我要写书评

暂无评论

Multi Meta Information Fusion Graph Neural Networks for Session-based Recommendation

Journal of Network Intelligence

引用

Journal of Network Intelligence 2025年第1期10卷 366-382页

作者： Wen, Yan Sun, Hui-Zheng Liu, Liang Zhou, Ya-Kun Bian, Wei Yan, Ming-Hai College of Computer Science and Engineering Shandong University of Science and Technology Qingdao266590 China Shandong Port Group Co. Ltd Qingdao266000 China Geography and Environment Management Faculty of Environment Universoty of Waterloo WaterlooN2L3G1 Canada

Session-based recommendation aims to predict the anonymous user’s next click action based on his/her latest click sequence. Most state-of-the-art approaches use graph neural networks to model sessions as graphs to capture dynamic transitions between items within a session. Inspired by recent progress on hypergraph-based neural networks, we think it is a good candidate to model the commonly related items within a session, and it can provide a sound complement to the normal graph models. In addition, cross session information may also be helpful. But the excessive noise introduced by other sessions is a big challenge and is still not be well resolved by existing works. To this end, we propose a new approach, called Multi Meta Information Fusion Graph Neural Networks (MM-GNN) to infer user preferences for the current session in a more comprehensive way for the anonymous *** fuse information from normal graph and hypergraph to fully exploit their capabilities of capturing sequential behavior and group intention within a session. In order to better use cross session information, we propose a globallevel subsequent item injection module, which injects information of the most possible subsequent item of the last item in the current session in the global scope. Besides, to learn richer information on the hypergraph structure, we model the whole session as a hyperedge to complement sliding windows-based hyperedges when creating the hypergraph. We have evaluated the method on several benchmark datasets and the results show that our method outperforms state-of-the-art methods. © 2025, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

MARM-HyLiWi: Mobility-Aware Rate Maximization in Link Aggregation Enabled Hybrid Li-Fi/Wi-Fi Networks 17

MARM-HyLiWi: Mobility-Aware Rate Maximization in Link Aggreg...

引用

17th International Conference on COMmunication Systems and NETworkS, COMSNETS 2025

作者： Paramita, Saswati Mondal, Rajorshi Bhattacharya, Arani Bohara, Vivek Ashok Srivastava, Anand Delhi Wirocomm Research Group Center of Excellence Li-Fi Department of Electronics and Communication Engineering New Delhi India IIIT Delhi Department of Computer Science and Engineering New Delhi India

ISBN: (纸本)9798331531195

Light Fidelity (Li-Fi) offers high data rates in confined areas but is limited by susceptibility to signal blockages and constrained coverage. To address these limitations, Li-Fi is often integrated with Wi-Fi in a heterogeneous network (HetNet), leveraging Li-Fi's speed with Wi-Fi's broader coverage. However, hybrid networks introduce frequent handovers between Li-Fi and Wi-Fi access points (APs), especially for users near blind spots or with limited mobility. This paper proposes a link aggregation (LA) approach for a Li-Fi and Wi-Fi HetNet (LA-HLWN), where users receive data from both types of APs simultaneously, enhancing data rates and reducing handover frequency. To manage network load effectively, we develop an algorithm for dynamic data rate allocation, formulated as a multi-dimensional commodity flow problem that considers factors such as user requirements, AP signal quality, and backhaul conditions. Given the NP-hard nature of this problem, our solution models it initially as an integer linear programming (ILP) challenge and then applies a relaxed linear programming (LPP) approach with a rounding technique for real-time allocation. Additionally, to balance throughput and reduce the computational overhead, we implement a strategy to control handover frequency and load distribution across APs, achieving a 54% improvement in throughput. This approach demonstrates the potential of LA in enhancing user satisfaction and system performance in hybrid Li-Fi/Wi-Fi networks. © 2025 IEEE.

关键词： Wireless local area networks (WLAN)

来源：评论

学校读者我要写书评

暂无评论

NEMO: ML based fish species classification 2

NEMO: ML based fish species classification

引用

2nd International Conference on Computational Intelligence, Communication Technology and Networking, CICTN 2025

作者： Singh, Manvendra Gaur, Sreesh Tyagi, Sarthak Sharma, Udit Yadav, Prince Mishra, Adarsh Department of Computer Science KIET Group of Institutions Delhi Ghaziabad India Department of CSE-AIML ABES Engineering College U.P. Ghaziabad India

ISBN: (纸本)9798331530389

This paper describes the creation of an ML-based tool for image-based identification of fish species commonly found in Indian oceans, addressing the issues that fishermen experience in accurately reporting their catch. Motivated by the Smart India Hackathon (SIH) issue statement from the Indian National Centre for Ocean Information Services (INCOIS), our goal is to improve marine fishing advice services. We developed a dataset that includes photos of 21 fish species typically captured by Indian fishermen, precisely chosen to ensure clarity and little background noise. Using this dataset, we applied transfer learning to eight different pre-trained models to determine the most effective one for deployment in a mobile application. The Kotlin-developed program has an easy-to-use interface with catch history, fish nutritional values, and pertinent weather data. Our findings show that the installed models can correctly recognize fish species, with MobileNetV2 reaching a peak accuracy of 97.61%. In addition to expediting the species identification process, this cutting-edge tool provides fishermen with crucial information, which eventually leads to more accurate species-level catch reporting. © 2025 IEEE.

关键词： Fish

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

Industrial Pose Detection System Based on FastDTW and MediaPipe 2nd

Industrial Pose Detection System Based on FastDTW and MediaP...

引用

2nd International Conference on the Frontiers of Robotics and Software engineering, FRSE 2024

作者： Huang, Baoyi Liao, Jiaxing Dai, Huawei Wang, Yuxuan School of Computer Science and Engineering Huizhou University Huizhou China Han’s Laser Technology Industry Group Co. Ltd. Shenzhen China

ISBN: (纸本)9789819631605

The posture and mobility of workers in industrial production directly affect productivity, safety, and product quality. However, existing posture monitoring technologies for large-scale applications and real-time monitoring are *** utilize an improved FastDTW and a gesture-to-video matching algorithm that normalizes feature vectors. An industrial posture detection system was developed to enhance the efficiency and quality of the production line, with a prototype based on the Orange Pi 3B. Our algorithm was compared to advanced gesture recognition algorithms such as OpenPose, DeepLabCut, and TensorFlow. The results show that our method exhibits higher accuracy and faster processing speed due to its efficient feature extraction and optimized similarity calculation. The optimized hardware design and operating system support enable rapid processing and analysis of real-time data streams, combining the advantages of geometric feature extraction and FastDTW. Compared to other components, our system surpasses existing gesture recognition methods in many ways, opening new possibilities for real-time applications. The accuracy rate in the test data reaches 90%, which is comparable to other algorithms but less computationally intensive. This automated posture detection system improves productivity and quality, bringing significant technological innovation to industrial assembly lines. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Digital storage

来源：评论

学校读者我要写书评

暂无评论

Comparison of Standard Ant Colony Optimization, Dijkstra and A∗ Algorithms for Path Planning of Robot in an Indoor Environment

Comparison of Standard Ant Colony Optimization, Dijkstra and...

引用

9th International Conference on Mechanical engineering and Robotics Research, ICMERR 2025

作者： Sharan, Sudeep Ngoc Minh, Anh Tong Dominguez-Jimenez, Juan Jose Nauth, Peter Frankfurt University of Applied Sciences Faculty of Computer Science and Engineering Frankfurt am Main Germany UCASE Software Engineering Group School of Engineering University of Cádiz Cadiz Spain

ISBN: (纸本)9798331509262

The research topic of Path planning is extremely challenging area of concentration within the field of mobile robots. However, path planning algorithms for mobile robot tasks are contingent upon the environment and its level of complexity. This paper analyzes four distinct path planning methods for simulated indoor environment. The proposed algorithms include conventional Ant Colony methods, Ant System (AS), Ant Colony System (ACS), as well as standard methods Dijkstra and AStar (A*). We analyzed and examined these algorithms by employing various metrics/maps with complexity. The results indicate that the traditional path planning algorithm Dijkstra and A∗ approaches surpass the other ant colony techniques in terms of both computation time and path distance. © 2025 IEEE.

关键词： Ant colony optimization

来源：评论

学校读者我要写书评

暂无评论

Uncertainty and Stochasticity in Modelling of Microstructure Evolution During Hot Rolling of Steels

引用

Archives of Metallurgy and Materials 2025年第1期70卷 405-414页

作者： Szeliga, D. Foryś, J. Kusiak, J. Kuziak, R. Nadolski, R. Pietrzyk, M. Rauch, L. AGH University of Science and Technology Faculty of Metals Engineering and Industrial Computer Science Al . Mickiewicza 30 Kraków30-059 Poland Research Group Processes Simulation Upper-Silesian Institute of Technology ul. K. Miarki 12-14 Gliwice44-100 Poland

Advanced numerical models, which predict heterogeneity of microstructural features, are needed to design modern steels with heterogeneous microstructures. Models based on stochastic internal variables meet this requirement. Our stochastic model accounts for the random character of the recrystallization and transfers this randomness into equations describing the evolution of the dislocation populations and the grain size during the hot deformation of steels. The idea of the internal variable model, with the dislocation density and the grain size being stochastic variables, is described in the paper. The material parameters, which influence accuracy and reliability of the model, were identified. They compose shear modulus, lattice friction stress and the mean free pass for dislocations. Numerical test showing influence of these parameters on the identification of the model coefficients were performed and a hint how these parameters should be selected is given. Compression loads and histograms of the grain size measured in the experimental compression tests were used to identify the coefficients in the model. The model was applied to simulations of the industrial process of the hot strip rolling. It was shown that the model can be used to both predictions of the microstructural heterogeneity caused by the stochastic character of microstructure evolution and to the evaluation of the uncertainty of phase composition in the final product. The latter is due to the uncertainty of the boundary conditions. © 2025. The Author(s).

关键词： Stochastic models

来源：评论

学校读者我要写书评

暂无评论

PES-DAS: Privacy Enhancing Subset-Data Aggregation Scheme Leveraging Homomorphic Cryptosystem in Fog-Cloud Based IoT Applications

引用

SN computer science 2025年第3期6卷 1-16页

作者： Shukla, Prashant Pandey, Sudhakar Department of Information Technology National Institute of Technology Chhattisgarh Raipur 492010 India School of Computer Science Engineering and Technology Bennett University (The Times Group) Uttar Pradesh Greater Noida India

Fog computing is an emerging technology that many sectors and industries are currently investing in to establish their own real-time and low-latency environments. Unlike the traditional Internet of Things (IoT), the fog-based IoT systems offers improved performance efficiency and enhanced data security by providing local mechanisms for pre-processing, filtering, and forwarding of data. However, the incorporation of fog technologies into the IoT has specific challenges with security and privacy. This is because fog nodes are located at the edge of the network, that may lack full reliability and trustworthiness. This work presents a novel scheme called PHE-SDA for aggregating a subset of response data while preserving privacy. The proposed PHE-SDA scheme is specifically designed for scenarios involving Fog-Cloud based IoT systems. It enables end-users to calculate the total amount of data collected from a certain group of smart devices. The subset is determined through securely calculating the inner product to measure the similarity between the normalized feature vectors of the end user's request and the response from each IoT device. If the inner product surpasses the threshold specified by the user, the data from the IoT device will be securely incorporated into the aggregation process to produce the ultimate outcome. In order to successfully carry out privacy-enhancing subset data aggregation in the Fog-Cloud based IoT system, we employ the Paillier homomorphic cryptography technique to encrypt the user's desired feature vector, similarity threshold, data sensed by IoT device, and some intermediate results. This work is an innovative initiative to effectively address the problem of prioritizing privacy while aggregating subsets in fog-based IoT systems. Subsequently, this work presents a comprehensive analysis and assessment of the effectiveness of the proposed PHE-SDA scheme. The comprehensive evaluation and results indicate that our proposed PHE-SDA scheme succes

关键词： Aggregation scheme Fog-based IoT network Homomorphic encryption Privacy preserving Subset data

来源：评论

学校读者我要写书评

暂无评论

A novel WO-ANT: whale-ant optimization algorithm for detection of Parkinson’s disease

引用

International Journal of Information Technology (Singapore) 2025年第3期17卷 1873-1881页

作者： Singh, Neha Sinha, Sapna Singh, Laxman Amity Institute of Information Technology Amity University Uttar Pradesh Noida India Department of Computer Science and Engineering (AI-ML) KIET Group of Institutions U.P Ghaziabad India

Parkinson’s disease (PD) is a neurological condition that impacts the quality of life for millions of people all over the world. A prompt and precise diagnosis of PD is absolutely necessary for the successful treatment and management of the condition. For the detection of PD, numerous methods exist in the literature, including speech-based methods, gait-based methods, and handwriting-based methods. Compared to the other two methods, the speech-based method is regarded as the most effective and competitive. The Whale-Ant Optimization Algorithm (WO-ANT) is a novel swarm optimization strategy that is presented in this paper as a potential method for the detection of PD via feature selection. To help in early detection, the proposed method attempts to enhance classification accuracy of PD detection by selecting an optimum subset of features from a wide feature space. In this paper, authors evaluate the classification efficacy of WO-ANT for PD using a speech-based method. MATLAB is used to implement the proposed method on the PD dataset obtained from ***. The efficacy of the algorithm is evaluated based on the optimal subset of features, accuracy, and computational cost. WO-ANT accurately reduced the number of features from 23 to 4 giving an accuracy of 90.29%. According to the numerical results, the algorithm performed well in terms of accuracy and selection of the optimal number of features. © Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： Ant colony optimization algorithm Feature selection Parkinson’s disease Whale optimization algorithm Whale-ant optimization algorithm

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：