检索结果-内蒙古大学图书馆

Very long wave infrared quantum dot photodetector up to 18μm

Light(science & Applications) 2024年第4期13卷 772-782页

作者： Xiaomeng Xue Qun Hao Menglu Chen School of Optics and Photonics Beijing Institute of TechnologyBeijing 100081China. Westlake Institute for Optoelectronics FuyangHangzhou 311421China. Physics Department Changchun University of Science and TechnologyChangchun 130022China

Colloidal quantum dots(CQDs)are of interest for optoelectronic devices because of the possibility of high-throughput solution processing and the wide energy gap tunability from ultraviolet to infrared *** may question about the upper limit on the CQD wavelength *** date,although the CQD absorption already reaches terahertz,the practical photodetection wavelength is limited within mid-wave *** figure out challenges on CQD photoresponse in longer wavelength,would reveal the ultimate property on these ***’s more,it motivates interest in bottom-up infrared photodetection with less than 10%cost compared with epitaxial growth semiconductor *** this work,developing a re-growth method and ionic doping modification,we demonstrate photodetection up to 18μm wavelength on HgTe *** liquid nitrogen temperature,the responsivity reaches 0.3 A/W and 0.13 A/W,with specific detectivity 6.6×108 Jones and 2.3×109 Jones for 18μm and 10μm CQD photoconductors,*** work is a step toward answering the general question on the CQD photodetection wavelength limitation.

关键词： quantum wave ultraviolet

来源：评论

学校读者我要写书评

暂无评论

A novel method for predicting the interaction between Chinese herbal compounds and targets 3

A novel method for predicting the interaction between Chines...

引用

3rd International Conference on Biomedical and Intelligent Systems, IC-BIS 2024

作者： Wang, Hongmei Liu, Geanqi Xu, Ming School of Computer Science & Engineering Changchun University of Technology Changchun130102 China

ISBN: (纸本)9781510681279

The active components and target points of traditional Chinese medicine are highly complex and difficult to ascertain. In recent years, computational methods have become an effective approach for predicting compound-target interactions. However, effective utilization of the topological structural features of the network remains an urgent challenge. Traditional prediction models tend to focus excessively on the relationships between nodes while overlooking the features of the edges. To address these issues, we propose a TCMCPI model to predict the interactions between herbal compounds and target points. First, we transform the CPI network into a line graph by establishing a topological structure of the edges. We then used GCN and CNN to extract features from the compounds and targets. These features are fused together to represent the nodes of the line graph, and GAT is utilized to learn the node features. Finally, we map the node features of the line graph back to the edge features in the CPI network by merging the adjacent edge features into node features. The experimental results demonstrate that TCMCPI outperforms other baseline methods in predicting interactions between herbal compounds and target proteins, and high AUROC and AUPR evaluation metrics attest to its superior predictive performance. © 2024 SPIE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Advancing Bankruptcy Forecasting With Hybrid Machine Learning Techniques: Insights From an Unbalanced Polish Dataset

引用

IEEE Access 2024年 12卷 9369-9381页

作者： Ainan, Ummey Hany Por, Lip Yee Chen, Yen-Lin Yang, Jing Ku, Chin Soon Universiti Malaya Faculty of Computer Science and Information Technology Department of Computer System and Technology Kuala Lumpur50603 Malaysia National Taipei University of Technology Department of Computer Science and Information Engineering Taipei106344 Taiwan Universiti Tunku Abdul Rahman Department of Computer Science Kampar31900 Malaysia

The challenge of bankruptcy prediction, critical for averting financial sector losses, is amplified by the prevalence of imbalanced datasets, which often skew prediction models. Addressing this, our study introduces the innovative hybrid model XGBoost+ANN, designed to leverage the strengths of both ensemble learning and artificial neural networks. This model integrates a comprehensive set of features with parameters optimized through genetic algorithms, eschewing traditional feature selection approaches. Our research focuses on an unbalanced dataset of Polish companies and reveals that the XGBoost+ANN model, in particular, exhibits outstanding performance. Optimized using genetic algorithms and without feature selection, this model achieved the highest AUC (0.958), sensitivity (0.752), and accuracy (0.983) scores, surpassing other models in our study. This remarkable outperformance, along with the robust results, marks a substantial advancement in the field of bankruptcy prediction. It underscores the efficacy of our approach in addressing the persistent challenge of data imbalance, offering a more reliable and accurate solution for financial risk assessment. © 2013 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Hopping-mean: an augmentation method for motor activity data towards real-time depression diagnosis using machine learning

引用

Multimedia Tools and Applications 2025年第18期84卷 18781-18799页

作者： Misgar, Muzafar Mehraj Bhatia, M.P.S. Department of Computer Science & Engineering Netaji Subhas University of Technology New Delhi India

The advances from the last few decades in the fields of ML (Machine Learning), DL (Deep Learning), and semantic computing are now changing the shape of the healthcare system. But, unlike physical health problems, diagnosis of mental health problems lacks a direct quantitative test. A huge variety of data generated by wearable IoMT (Internet of Medical Things) devices (e.g. wrist actigraph) provides an efficient objective way to diagnose such mental health problems in the early stages. In this paper, we propose a novel overall methodology comprised of preprocessing, data augmentation, and classification of depressive episodes from motor activity data. A time-efficient Augmentation Algorithm is proposed based on windowing over the same timestamps to resolve the problem of data loss in the Depresjon database and the UMAP (Uniform Manifold Approximation and Projection) based feature extraction method is used on motor activity database containing variable-length activity records for 23 control patients and 32 condition groups. The classification of depression from 24 h of motor activity data with augmentation performs much better than raw data. The preprocessing approach followed by Random forest provides cut over cut over the current state of the art and provides the highest accuracy of 74.29%, specificity of 74.80%, and sensitivity of 73.56 on non-augmented data and gives the highest accuracy of 98.65%, specificity of 97.36% and sensitivity of 99.89% on augmented data. This shows the effectiveness of using the custom augmentation approach proposed in this paper. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Medical problems

来源：评论

学校读者我要写书评

暂无评论

Grouped Federated Learning Algorithm Based on Non-IID Data 4

Grouped Federated Learning Algorithm Based on Non-IID Data

引用

4th International Conference on Machine Learning and computer Application, ICMLCA 2023

作者： Li, Zhongxin Zhang, Jianfei School of Computer Science and Technology Changchun University of Science and Technology China

ISBN: (纸本)9798400709449

Federated learning is a new machine learning paradigm in which multiple clients collaborate to train a machine learning model while protecting local data privacy. Client-side non-identically and non-independently distributed (Non-IID) data is one of the most important challenges for federated learning algorithms. The classification accuracy of the federated learning model drops drastically due to the weight dispersion caused by data heterogeneity. In this work, we adopt a way of dividing client data that is closer to real applications. We propose a federated learning algorithm based on rearranging all clients(FedRC). FedRC sets up a novel within-group training approach and a server-side pre-aggregation policy to address the impact of Non-IID on the model. In addition, we test FedRC on a simulated dataset and the experimental results demonstrate that FedRC outperforms the baseline algorithm and significantly improves the performance on Non-IID data. © 2023 ACM.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Federated Learning Based on Feature Fusion 4

Federated Learning Based on Feature Fusion

引用

4th International Conference on Neural Networks, Information and Communication Engineering, NNICE 2024

作者： Cao, Junjie Chen, Zhiyu Changchun University of Technology School of Computer Science & Engineering Changchun130102 China

ISBN: (纸本)9798350394375

Federated learning is a distributed machine learning method that has important research significance in areas such as privacy, data heterogeneity, communication efficiency, and incentive mechanisms. Data imbalance is a key challenge faced by federated learning. Due to the non iid nature of the data provided by each client in federated learning, the client is unable to obtain a good local model during client training. The iterative process of uploading, aggregating, and downloading the model hinders its ability to make accurate predictions. In order to solve the problem of data imbalance, this paper constructs multiple auxiliary branches in the local model and fuses their features to improve the prediction accuracy of the model in the case of data imbalance. The experimental results show that the model has improved performance in both scenarios of simulating 100 and 500 clients. © 2024 IEEE.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Trademark detection and classification based on an improved YOLO

引用

Multimedia Tools and Applications 2025年 1-17页

作者： Miao, Chunyuan Wang, Xiuhui Department of Computer Science and Technology China Jiliang University Zhejiang Hangzhou310018 China

Trademarks are important identifiers for goods and services. They play an increasingly important role in daily life and production. However, with the continuous development of commercial society and the increasing number of goods and services, trademark detection and classification technology has become particularly critical, and it has large application prospects in the fields of commodity image search and trademark infringement. Based on the You Only Look Once (YOLO)-v5 framework, this paper proposes the YOLO-BC network to better detect trademark images. First, the network adopts a new bi-directional feature pyramid network & coordinate attention(BiFC) module to replace the original enhanced feature extraction network (PANet) model to improve the feature extraction ability of the network at different scales and capture key information. Second, a convolutional block attention module (CBAM) is introduced into the C3 module, and the influence of the C3-CBAM module embedded in the backbone and neck of the algorithm network on network performance is explored through experiments. Subsequently, the distance intersection over union non-maximum suppression (DIOU-NMS) is used to replace the NMS in the original network as the bounding box non-maximum suppression method. Finally, we replace generalized IOU (GIOU) with α-IOU as the bounding box regression loss function and use the appropriate parameter α to adaptively adjust the loss and gradient of the IOU target. Experimental results on public dataset and a self-constructed dataset show that the improved framework has a higher detection accuracy than traditional methods. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Research on General-Purpose Brain-Inspired Computing Systems

引用

Journal of computer science & technology 2024年第1期39卷 4-21页

作者：渠鹏纪兴龙陈嘉杰庞猛李宇晨刘晓义张悠慧 Department of Computer Science and Technology Tsinghua UniversityBeijing 100084China Department of Precision Instrument Tsinghua UniversityBeijing 100084China

Brain-inspired computing is a new technology that draws on the principles of brain science and is oriented to the efficient development of artificial general intelligence(AGI),and a brain-inspired computing system is a hierarchical system composed of neuromorphic chips,basic software and hardware,and algorithms/applications that embody this *** the system is developing rapidly,it faces various challenges and opportunities brought by interdisciplinary research,including the issue of software and hardware *** paper analyzes the status quo of brain-inspired computing *** by some design principle and methodology of general-purpose computers,it is proposed to construct"general-purpose"brain-inspired computing systems.A general-purpose brain-inspired computing system refers to a brain-inspired computing hierarchy constructed based on the design philosophy of decoupling software and hardware,which can flexibly support various brain-inspired computing applications and neuromorphic chips with different ***,this paper introduces our recent work in these aspects,including the ANN(artificial neural network)/SNN(spiking neural network)development tools,the hardware agnostic compilation infrastructure,and the chip micro-archi-tecture with high flexibility of programming and high performance;these studies show that the"general-purpose"system can remarkably improve the efficiency of application development and enhance the productivity of basic software,thereby being conductive to accelerating the advancement of various brain-inspired algorithms and *** believe that this is the key to the collaborative research and development,and the evolution of applications,basic software and chips in this field,and conducive to building a favorable software/hardware ecosystem of brain-inspired computing.

关键词： brain-inspired computing neuromorphic chip compiler spiking neural network

来源：评论

学校读者我要写书评

暂无评论

Optimal Design and Mechanical Simulation of Rubber bushing with Convex Hull Structure Based on Bionics

引用

Journal of Bionic Engineering 2023年第5期20卷 2018-2029页

作者： Ce Liang Min Li Jicai Liang Shaoqiang Wang Qigang Han Yi Li Key Laboratory of Automobile Materials Ministry of Educationand College of Materials Science and EngineeringJilin UniversityChangchun130025China Roll Forging Research Institute Jilin UniversityChangchun130025China School of Computer Science and Technology Changchun UniversityChangchun130022China

Inspired by the safe landing of cats falling from high altitudes,a bionic flexible rubber bushing structure is proposed and its motion characteristics are systematically studied to explore its potential application in the suppression of *** convex hull structure on the bushing surface is abstracted from the cat’s claw pad,and the hyper-viscoelastic model is selected as the constitutive model of the rubber *** addition,the design with the best vibration damping effect is finally obtained by reasonably adjusting the amount of radial compression and distribution of bionic ***,under the same conditions,the test results of the dynamic characteristics of the bushing verify the accuracy of the simulation *** results show that the convex hull bionic structure designed in this paper can effectively change the motion characteristics of the rubber bushing under various working conditions,which provides new inspiration or potential possibility for the design of rubber bushing in the future.

关键词： Rubber bushing Bionic design Finite element analysis Vibration isolator Radial compression

来源：评论

学校读者我要写书评

暂无评论

Underwater object detection based on enhanced YOLOv4 architecture

引用

Multimedia Tools and Applications 2024年第18期83卷 53759-53783页

作者： Liu, Ching-Hua Lin, Chang Hong Department of Electronic and Computer Engineering National Taiwan University of Science and Technology Taiwan

Object detection and image restoration pose significant challenges in deep learning and computer vision. These tasks are widely employed in various applications, and there is an increasing demand for specialized environments where images are prone to blur or noise, which can adversely affect subsequent results. In recent years, significant breakthroughs have been achieved in object detection performance. While some previously proposed methods prioritize high accuracy at the cost of longer inference times, others prioritize speed. Therefore, it is crucial to design an efficient network architecture that maintains both inference speed and high accuracy. This research proposes a network architecture for underwater object detection with an attention mechanism. The proposed approach differentiates itself from other methods by employing a deblurring network as a preprocessing step to restore and enhance the image quality of the underwater dataset. Additionally, in the feature extraction stage of the detection network, channel and spatial feature information are individually enhanced. These adaptive attention features are then integrated into a multi-scale feature fusion. Finally, the cross-stage local method is combined to improve the learning ability of the convolutional neural network while reducing the size of the model. Based on the experimental results, our proposed model architecture achieves leading accuracy and strikes a favorable balance in terms of model size compared to previously proposed methods. Based on our structure, the metrics of AP and AP50 reach 66.8 and 87.6, respectively. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2023.

关键词： Image reconstruction

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：