检索结果-内蒙古大学图书馆

2nd International Conference on Machine Learning and Data engineering, ICMLDE 2023

作者： Syed, Liyakathunisa Department of Computer Science College of Computer Science and Engineering Taibah University Madinah Saudi Arabia

Crop yield is a serious concern for farmers due to irregular irrigation, soil erosion, uncontrolled seed planting, severe weather conditions, peasants, and unpredicted locust infestations. The lack of current data and the intricacy of traditional agriculture leads to inefficiency and excessive operation costs. Smart farming with Artificial Intelligence techniques and the Internet of Things (IoT) in agriculture overcome these challenges by connecting all the accessible data sources into a single, fully effective functional unit. Smart agriculture helps in the sustainability of food production by using minimum resources such as water, fertilizer, and seeds. It provides a better understanding of the soil, crop, and changing weather. Furthermore, the sensors in the system assist in monitoring and controlling the resources. In this research, we present ensemble-based machine learning approaches in the IoT environment to predict crop yields and enable sustainable farming by guiding the farmers to grow the correct crop during the correct season to increase their yields. The proposed method uses a novel ensemble-based machine learning classification approach with two levels of predictions-level-0 prediction includes (Logistic Regression (LR), Classification and Regression Trees (CART), Support Vector Machines (SVM), and K-Nearest Neighbors (KNN)) classifiers are input as features to the level-1 meta classifier (Random Forest) to detect the distinct categories of different crops. Its predictions can be extremely close to the ground truth, which accurately predicts 22 categories of crops with 99% accuracy. Hence, the proposed Smart agriculture system can assist farmers in improved yield production and remote monitoring at low cost. © 2024 Elsevier B.V.. All rights reserved.

关键词： Crops

来源：评论

学校读者我要写书评

暂无评论

A Survey on Task Scheduling in Edge-Cloud

引用

SN computer science 2025年第3期6卷 1-21页

作者： Sahoo, Subham Kumar Mishra, Sambit Kumar Department of Computer Science and Engineering SRM University AP Amaravati India

In this modern era, cloud computing is not enough to meet today’s intelligent society’s data processing needs, so edge computing has emerged. In contrast to computation in the cloud, it elaborates user proximity and proximity to the data source. To store local, small sized, and processed data on the edges of the network is more effective. The edge paradigm, intended to be a leading computation due to its low latency, also faces many challenges due to computational capabilities and resource availability. Edge computing allows edge devices to release heavy loads and computational operations on the remote server. This allows us to take full advantage of the server-side computing and storage in edge devices. However, the offload of all highly compressed computing operations on a remote server at the same time may become overcrowded, leading to intensive processing delays for many computing operations and unexpectedly elevated power usage. Instead of that, it is possible that spare edge resources may need to be utilized effectively and the access to expensive cloud resources would be restricted. As a result, it is important to investigate the collaborative planning process (scheduling) for the edge servers with a cloud server based on task features, development objectives, and system status. It can assist in performing all the computing functions efficiently and effectively. This paper analyzes and summarizes computing conditions for the edge computing context and classifies the computation of tasks into various edge-cloud computing scenarios. At the end, based on the problem structure, various collaborative planning methods for computational functions are presented. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Cloud computing Edge computing Energy consumption Makespan Task scheduling

来源：评论

学校读者我要写书评

暂无评论

Skyline Frequent-utility Patterns Mining: A Survey

Journal of Network Intelligence

引用

Journal of Network Intelligence 2023年第1期8卷 1-16页

作者： Wu, Jimmy Ming-Tai Zhou, Huiying Pirouz, Matin Tayeb, Shahab College of Computer Science and Engineering Shandong University of Science and Technology Qindao China Department of Computer Science California State University Fresno California United States Department of Electrical and Computer Engineering California State University Fresno California United States

High Utility Itemset Mining (HUIM) and Frequent Itemset Mining (FIM) are two important branches in the data mining area, where Frequent Itemset Mining considers itemsets that occur in large numbers in the transaction database, while High Utility Itemset Mining considers the number and unit utility value of itemsets. However, in practical applications, both mining approaches have their own limitations, for example, itemsets with a high number of occurrences may contribute less to the ultimate goal of the merchant, and although these itemsets are strongly related to each other, they do not have great reference value for the final decision. The itemsets with high utility values, on the other hand, may have a very low number of occurrences and weak connections between the itemsets. Therefore recommending these itemsets to users may not be very meaningful. In order to better meet the requirements of real life, some scholars took these two mining algorithms into account, hoping to find those high-quality itemsets, that is, those itemsets with more occurrences and high utility values. For this purpose, the skyline frequent-utility pattern mining (SFUPM) algorithm is introduced. This paper describes in detail the classification of skyline frequent-high uyility itemset mining, mining methods and future research directions. © 2023, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Introducing Learning Rate Adaptation CMA-ES into Rigid Intraoperative 2D/3D Registration for Spinal Surgery 22

Introducing Learning Rate Adaptation CMA-ES into Rigid Intra...

引用

22nd IEEE International Symposium on Biomedical Imaging, ISBI 2025

作者： Chen, Minheng Zhang, Zhirun University of Texas Department of Computer Science and Engineering Arlington United States School of Computer Science and Engineering Southeast University China

ISBN: (纸本)9798331520526

The covariance matrix adaptive evolution strategy (CMA-ES) has been widely used in the field of 2D/3D registration in recent years. This optimization method exhibits exceptional robustness and usability for complex surgical scenarios. However, due to the inherent ill-posed nature of the 2D/3D registration task and the presence of numerous local minima in the landscape of similarity measures, evolution strategies often require a larger population size in each generation to ensure the stability of registration and the globality and effectiveness of search, which makes the entire process computationally expensive. In this paper, we build a 2D/3D registration framework based on a learning rate adaptation CMA-ES manner. The framework employs a fixed and small population size, leading to minimized runtime and optimal utilization of computing resources. We conduct experimental comparisons between the proposed framework and other intensity-based baselines using a substantial volume of synthetic data. The results suggest that our method demonstrates superiority in both registration accuracy and running time compared with other optimization-based baselines. Code is available at ***/m1nhengChen/CMAES-reg. © 2025 IEEE.

关键词： 2D/3D registration CMA-ES Evolutionary computing Image-guided interventions

来源：评论

学校读者我要写书评

暂无评论

Optimization-Driven Vendor Selection with Tabu Search and Neural Networks 14

Optimization-Driven Vendor Selection with Tabu Search and Ne...

引用

14th IEEE International Conference on Communication Systems and Network Technologies, CSNT 2025

作者： Gupta, Santosh Kr. Srivastava, Anubhava Kumar, Vivek Department of Computer Science and Engineering SSET Sharda University Greater Noida India Department of Computer Science and Engineering NIET Greater Noida India

ISBN: (纸本)9798331531935

Vendor selection in supply chain management is a complex but extremely important process that requires the evaluation of multiple factors such as cost, quality, delivery time, and responsiveness. This study presents a new hybrid methodology that makes the vendor selection process more efficient by combining tabu search optimization and neural network analysis. The proposed model operates in two stages. In the first stage, tabu search algorithms systematically analyze the solution space and identify optimal vendor combinations based on predefined criteria. In the second stage, these selected combinations are evaluated through a neural network. This trained network has the ability to recognize patterns in performance indicators such as cost efficiency, delivery time, and operational wastage. Tests conducted on simulated data clearly demonstrated the superior performance of this hybrid model compared to traditional single-method approaches. The model significantly improved key parameters such as Mean Absolute Error and R2 score. Moreover, it also proved to be adaptable to changing conditions. This framework not only addresses challenges such as scalability and integration of qualitative-quantitative factors in vendor selection, but also highlights the effectiveness of combining optimization algorithms and machine learning techniques for complex supply chain decisions. This model not only serves as a practical decision-support tool for procurement managers but also lays a strong foundation for the development of hybrid AI techniques in the field of operational optimization. © 2025 IEEE.

关键词： Tabu search

来源：评论

学校读者我要写书评

暂无评论

Sensor Switching-Based Automatic Misalignment Detection and Correction System for Wireless Power Transfer

IEEE Open Journal of Vehicular Technology

引用

IEEE Open Journal of Vehicular Technology 2025年 6卷 1380-1398页

作者： Dahhan, Ali Younis Al Mukhopadhyay, Shayok Hassan, Mohamed S. Osman, Ahmed H. American University of Sharjah Department of Electrical Engineering P.O. Box 26666 Sharjah United Arab Emirates Electrical & Computer Engineering and Computer Science Department West HavenCT06516 United States

Misalignment between the transmitting and receiving coils is an inevitable problem for electric vehicle (EV) wireless power transfer (WPT) systems. Regardless of the WPT system being static or dynamic, coil misalignment reduces the efficiency of the charging system. This paper, focuses on using a combination of computer vision and one of two different misalignment sensors to detect, and further correct lateral misalignment between the EV receiving (Rx) coil and the segmented transmitting (Tx) coils in a charging lane. The vision-based component uses a camera for lane detection and is primarily responsible for detecting larger deviations and making coarse compensations by estimating the lateral shift of the EV, relative to the center of the charging lane. The sensor-based approach relies on Hall effect sensors or detection coils to detect the misalignment in a smaller range, and perform finer corrections. A one-dimensional (1D) actuator moves the receiving coil to correct the coil misalignment, independent of vehicle movements. The vision-based approach showed a wide detection range for misalignment spanning [−15,15] cm, with a correction accuracy of ≈ ±2 cm. This is juxtaposed with the sensor-based approach which operates on a misalignment range of [−3,3] cm, but outperforms the vision-based approach with a correction accuracy of less than ±1 mm. The proposed sensor switching-based approach combines the advantages of the above individual techniques. An experimental setup is developed and tests are performed to evaluate the proposed approach while transferring 108 W of power wirelessly © 2020 IEEE.

关键词： Coils Switches Hall effect Wireless power transfer Inductance Estimation Electric vehicles Costs Batteries Anxiety disorders

来源：评论

学校读者我要写书评

暂无评论

Bone Osteoporosis Fractures Detection with Deep Learning: An X-ray Image Analysis Approach

Biomedical Materials and Devices

引用

Biomedical Materials and Devices 2025年 1-19页

作者： Belali, Md. Habibullah Khan, Ferdus Rhaman Rajin, Nur Mohammed Sarkar, Hridoy Uddin, Khandaker Mohammad Mohi Department of Computer Science and Engineering Southeast University Dhaka 1208 China

This research aims to develop a CNN-based computational framework for diagnosing bone fractures from X-rays, improving osteoporosis knowledge. Current osteoporosis management focuses on elderly fracture prevention, but many are unaware of their condition until a fracture occurs. This framework seeks to enable earlier diagnosis and management for all individuals, preventing fractures and improving quality of life, especially for at-risk and aging populations. These interesting findings prove modern deep learning systems as potential diagnoses for bone fractures even in cases of ones suffering from osteoporosis. This paper examines the use of DEXA radiography augmented by deep learning algorithms to estimate fracture risk beyond mineral density determination. The existence of computer-aided designs in numerous areas cuts across and helps to address the issues that are presented with the current methods of diagnosis. In the field of bone radiography, X-ray has been opted for in this study since it is more effective than CT scans. In this research, an advanced CNN model was applied that included a mix of architectural modifications to classical models and data augmentation techniques to outperform other standard CNN-based models. The Canny edge detection algorithm and the contour detection algorithm, both based on OpenCV, were incorporated and supplemented using Adam optimization. The focus of this work has been the enhancement of CNN model to detect fractures and localize it at the same time. The best model solely attained an accuracy level of 98.23%, outperforming all other tested models, including VGG16, VGG19, Random Forest Classifier, ViT’s, and its various versions. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： BMD CNN DEXA LeNet-5 Osteoporosis ViT’s X-ray

来源：评论

学校读者我要写书评

暂无评论

Unlocking Eye Gestures with Earable Inertial Sensing for Accessible HCI 17

Unlocking Eye Gestures with Earable Inertial Sensing for Acc...

引用

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

作者： Chugh, Garvit Chakraborty, Suchetana Chakraborty, Sandip Indian Institute of Technology Department of Computer Science and Engineering Jodhpur India Indian Institute of Technology Department of Computer Science and Engineering Kharagpur India

ISBN: (纸本)9798331531195

Eye gestures are widely used in many applications, including device control, biometrics, visual analytics, and health-care, like Alzheimer's, accessibility, etc. The conventional method for eye gesture detection needs visual information processing devices or smart glasses to capture the movement patterns of the eyeballs. In this paper, we present a novel framework that utilizes inertial measurement unit (IMU) sensors embedded in earable devices for real-time eye gesture detection, which can work as a complementary to visual information processing devices. The framework translates specific eye movements into actionable commands, enabling hands-free interaction for users with limited mobility. Through extensive experimentation with 14 participants across diverse activities such as sitting, walking, running, and driving, the framework demonstrated high accuracy in eye gesture recognition, achieving a macro F1 score of 0.85 on our self-collected dataset. © 2025 IEEE.

关键词： Eye movements

来源：评论

学校读者我要写书评

暂无评论

Drug–Target Interaction Prediction Model Using Optimal Recurrent Neural Network

引用

Intelligent Automation & Soft Computing 2023年第2期35卷 1675-1689页

作者： G.Kavipriya D.Manjula Department of Computer Science and Engineering Anna UniversityChennai600025India

Drug-target interactions prediction(DTIP)remains an important requirement in theﬁeld of drug discovery and human *** identiﬁcation of interaction among the drug compound and target protein plays an essential pro-cess in the drug discovery *** is a lengthier and complex process for pre-dicting the drug target interaction(DTI)utilizing experimental *** resolve these issues,computational intelligence based DTIP techniques were developed to offer an efﬁcient predictive model with low *** recently devel-oped deep learning(DL)models can be employed for the design of effective pre-dictive approaches for *** this motivation,this paper presents a new drug target interaction prediction using optimal recurrent neural network(DTIP-ORNN)*** goal of the DTIP-ORNN technique is to predict the DTIs in a semi-supervised way,i.e.,inclusion of both labelled and unlabelled ***,the DTIP-ORNN technique performs data preparation process and also includes class labelling process,where the target interactions from the database are used to determine theﬁnal label of the unlabelled ***,drug-to-drug(D-D)and target-to-target(T-T)interactions are used for the weight initia-tion of the RNN based bidirectional long short term memory(BiLSTM)model which is then utilized to the prediction of *** hyperparameters signiﬁ-cantly affect the prediction performance of the BiLSTM technique,the Adam optimizer is used which mainly helps to improve the DTI prediction *** order to ensure the enhanced predictive outcomes of the DTIP-ORNN techni-que,a series of simulations are implemented on four benchmark *** comparative result analysis shows the promising performance of the DTIP-ORNN method on the recent approaches.

关键词： Drug target interaction deep learning recurrent neural network parameter tuning semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

QoE-Aware Volumetric Video Caching and Rendering for Mobile Extended Reality Services

引用

IEEE Internet of Things Journal 2025年第12期12卷 21852-21865页

作者： Pei, Yingying Li, Mushu Huang, Xinyu Shen, Xuemin University of Waterloo Department of Electrical and Computer Engineering WaterlooONN2L 3G1 Canada Lehigh University Department of Computer Science and Engineering BethlehemPA18015 United States

In this article, we propose a novel volumetric video caching and rendering approach for an edge-assisted extended reality (XR) system to enhance user Quality of Experience (QoE). Particularly, user QoE consists of visual quality and quality variation. Different quality of volumetric videos are required to be cached, rendered, and delivered to XR devices for different viewing distances within a time latency. Given the limited caching, computing, and communication resources on the edge server, we formulate a long-term user QoE maximization problem to jointly optimize video caching and rendering by considering user locations and viewing distances. To solve this problem, we first design an online optimization algorithm in which caching decisions are obtained using a regularization technique. We then develop a low-complexity binary search algorithm to determine optimal rendering quality. Extensive simulations are conducted to demonstrate that our proposed approach outperforms benchmark schemes by an average 46% improvement in terms of long-term user QoE. © 2014 IEEE.

关键词： Video streaming

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：