检索结果-内蒙古大学图书馆

38th Conference on Neural information Processing Systems, NeurIPS 2024

作者： Liu, Mingyang Chen, Xinyang Shu, Yang Li, Xiucheng Guan, Weili Nie, Liqiang School of Computer Science and Technology Harbin Institute of Technology Shenzhen China School of Data Science and Engineering East China Normal University China School of Electronics and Information Engineering Harbin Institute of Technology Shenzhen China

Unsupervised domain adaptation excels in transferring knowledge from a labeled source domain to an unlabeled target domain, playing a critical role in time series applications. Existing time series domain adaptation methods either ignore frequency features or treat temporal and frequency features equally, which makes it challenging to fully exploit the advantages of both types of features. In this paper, we delve into transferability and discriminability, two crucial properties in transferable representation learning. It's insightful to note that frequency features are more discriminative within a specific domain, while temporal features show better transferability across domains. Based on the findings, we propose Adversarial CO-learning Networks (ACON), to enhance transferable representation learning through a collaborative learning manner in three aspects: (1) Considering the multi-periodicity in time series, multi-period frequency feature learning is proposed to enhance the discriminability of frequency features;(2) Temporal-frequency domain mutual learning is proposed to enhance the discriminability of temporal features in the source domain and improve the transferability of frequency features in the target domain;(3) Domain adversarial learning is conducted in the correlation subspaces of temporal-frequency features instead of original feature spaces to further enhance the transferability of both features. Extensive experiments conducted on a wide range of time series datasets and five common applications demonstrate the state-of-the-art performance of ACON. Code is available at https://***/mingyangliu1024/ACON. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Unstructured Healthcare Statistics Exploration Consuming Modest Ensemble Machine Learning Methods

Unstructured Healthcare Statistics Exploration Consuming Mod...

引用

2023 International Conference on Recent Advances in Science and engineering technology, ICRASET 2023

作者： Gunavathy, K. Agarwal, Ankita Sharma, Nipun Department of Computer Science and Information Technology Bangalore India Maharishi University of Information Technology Maharishi School of Engineering & Technology Uttar Pradesh India Presidency University Department of Electronics and Communication Engineering Karnataka Bangalore India

ISBN: (纸本)9798350306927

Various approaches to classifying emotions can be found inside an ML study. Of course, there are ML algorithms that do exceptionally well on particular text datasets but underwhelm on other, larger datasets. To get over this issue, several classifiers should be combined to boost classification performance across all datasets. In order to select the most effective classifiers for use in an ensemble, this study presents a competitive ensemble classification technique. The effectiveness of sentiment classification should improve as a result of this. Unstructured healthcare data is discussed together with the proposed competitive ensemble classification algorithm. This method employs Twitter data extraction to do multiclass sentiment categorization. They have been categorised into one of five mood groups. In this study, we propose a model that uses a competition between several types of classifiers to determine which ones will work best for ensemble learning. To enhance classification performance, weights are assigned to five separate classifiers, and both the suggested model and a benchmark are assessed. The proposed approach, based on empirical evidence, outperforms baseline categorization methods. To evaluate the effectiveness of the work proposed, experimental results are compared to standard datasets. The outcomes show that the proposed model is superior, with an increase in classification accuracy. © 2023 IEEE.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Optimal Visualization of Hyper Spectral Time Series Algorithms via Graphical Representations 5th

Optimal Visualization of Hyper Spectral Time Series Algorith...

引用

5th International Conference on Data Science, Machine Learning and Applications, ICDSMLA 2023

作者： Agarwal, Trapty Karthikeyan, M.P. Adhikari, Manoj S. Ulla, Mohammed Mujeer Maharishi School of Engineering & Technology Maharishi University of Information Technology Uttar Pradesh Lucknow India Karnataka Bangalore India Department of Computer Science & Engineering School of Engineering and Computing Dev Bhoomi Uttarakhand University Navgaon Manduwala Dehradun India School of Computer Science and Engineering Presidency University Karnataka Bengaluru India

ISBN: (纸本)9789819780426

This technical summary describes how using graphical representations can allow for the best visualization of hyper spectral time series algorithms. Hyper spectral time collection algorithms correctly analyze temporal information from satellite- or plane-borne sensors, for instance, to offer an in-intensity expertise of an environment or system. But, such evaluation can regularly be difficult to realize without graphical aid. This research gives a method for correctly visualizing hyper spectral time series algorithms using graphical representations, consisting of line graphs and tree diagrams. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Visualization

来源：评论

学校读者我要写书评

暂无评论

Hyper Spectral Image Analysis Integrating Fuzzy C-Means Clustering and Feature Selection 5th

Hyper Spectral Image Analysis Integrating Fuzzy C-Means Clus...

引用

5th International Conference on Data Science, Machine Learning and Applications, ICDSMLA 2023

作者： Agarwal, Trapty Rengarajan Goyal, Rohit Khan, N. Zafar Ali Maharishi School of Engineering & Technology Maharishi University of Information Technology Uttar Pradesh Lucknow India Karnataka Bangalore India Department of Computer Science & Engineering School of Engineering and Computing Dev Bhoomi Uttarakhand University Navgaon Manduwala Dehradun India School of Computer Science and Engineering Presidency University Karnataka Bengaluru India

ISBN: (纸本)9789819780426

Hyper spectral picture evaluation is an effective tool for analyzing large volumes of excessive-dimensional facts. In this paper, a brand new method to hyper spectral photo analysis is provided that integrates fuzzy C-means clustering and function choice. The proposed approach permits for significant quantification of the exclusive additives within a hyper spectral image and allows for the identification of key spectral functions. Additionally, the fuzzy C-way clustering approach is used to identify awesome clusters within the photograph, supplying a useful method for studying complicated datasets. The characteristic choice technique facilitates to further lessen the data set and to achieve the maximum meaningful capabilities for further evaluation. The overall performance of the proposed technique is evaluated via the use of an actual-international hyper spectral photo, and the effects exhibit the effectiveness of the proposed technique for hyper-spectral photograph analysis. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Image analysis

来源：评论

学校读者我要写书评

暂无评论

Improving Online Job Authenticity Detection Using Deep Learning and Focal Loss 2

Improving Online Job Authenticity Detection Using Deep Learn...

引用

2nd IEEE International Conference on Data Science and Network Security, ICDSNS 2024

作者： Akhila, K. Sofi, Mukhtar Ahmad Reddy, Ch. Likitha Sarayu, A. Sannitya, M. Bvrit Hyderabad College of Engineering for Women Department of Information Technology Hyderabad India Bvrit Hyderabad College of Engineering for Women Department of Computer Science Hyderabad India

ISBN: (纸本)9798350373110

The rise of fake job postings in online job marketplaces presents a significant challenge for job seekers, necessitating robust detection methods. Existing methods often struggle to adapt to evolving fraudulent tactics, leading to suboptimal accuracy. This project introduces a novel Deep Learning approaches which are Recurrent Neural Network (RNN) variants leveraging Bi-LSTM and Bi-GRU architectures to enhance fake job detection and used a loss function for removing the imbalanced class distribution which is called 'Focal Loss'. Using the EMSCAD dataset, we trained and evaluated the performance of the proposed model against state-of-the-art methods, aiming for broader accessibility through integration into a user-friendly webserver for real-time job posting analysis which will help jobseekers to identify whether the job posting is Genuine/Fake. In this project we got accuracy of value 98.6% with precision of 85%, recall of 84% and f1-score of 85%. © 2024 IEEE.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Vehicle Registration Plate Recognition Using YoloV8 and IOT 1

Vehicle Registration Plate Recognition Using YoloV8 and IOT

引用

1st International Conference on AIML-Applications for engineering and technology, ICAET 2025

作者： Nadaf, Juber Kadam, Amol Jadhav, Pramod Kadam, Prasad Patil, Vinod Gaikwad, Milind College of Engineering Department of Computer Pune India College of Engineering Department of CSBS Pune India College of Engineering Department of E & TC Pune India College of Engineering Department of Information Technology Pune India

ISBN: (纸本)9798350355611

Number Plate recognition or the vehicle License Plate recognition is an interesting topic of research in smart cities which utilizes Image Processing techniques. Since the number of cars is rising at an exponential rate, automated systems are supposed to store vehicle data for a variety of applications. In this intended algorithm, an efficient technique has been identified for recognition of different vehicle number registration plates. Through this technique, we can easily deal with problems like low light, irregular font number plates, wrong angles. We use the method of Optical Character Recognition to identify characters that are extracted from the registration plate. Even though this project works efficiently for common use cases, it has its own drawbacks and limitations even though we are not using deep learning or any other machine learning algorithms. One of the important steps in vehicle plate recognition is the character recognition which is a technique to obtain individual text characters. YOLOv8 (You Only Look Once), and Darknet- 53 are the methods that is used to extract the features of the data. In this model, the input data is the image of the vehicle that is captured from the camera sensors or cell phones and later cropped to display only the registration plate. We undergo data-pre-processing to improve the quality or the pixels of image on the number plate. Testing is done using two different models one is the model which is obtained with more pre-processing data and the other obtained without any pre-processing data. We evaluate two datasets of number plate images: one with no interference and one where the colour intensity of the image has been decreased to test, the accuracy of the model © 2025 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Urban green space extraction from BJ-2 remote sensing image based SegFormer semantic segmentation model (vol 51,2439835 ,2025)

引用

CANADIAN JOURNAL OF REMOTE SENSING 2025年第1期51卷

作者： Guo, Yan Shao, QuanSen Liu, FuJiang Lin, Weihua Wang, Mianzhi Li, Zhuowu Chen, Shuo Liu, Hongchen Su, Junshen Wang, Xianbin School of Computer Science China University of Geosciences (Wuhan) Wuhan China School of Geography and Information Engineering China University of Geosciences (Wuhan) Wuhan China China Geological Survey Xining Comprehensive Survey Center of Natural Resources Xining China Piesat Information Technology Co. Ltd Beijing China

Accurate extraction of urban green spaces is of great significance for promoting urban planning and management, and promoting sustainable urban development. Given the characteristic spatial distribution of urban green spaces in high-resolution remote sensing imagery, a framework for automated extraction of urban green spaces using the SegFormer semantic segmentation network is proposed. The framework leverages self-attention mechanisms to capture global information and enhance the extraction of green spaces. We have constructed an urban green space dataset that includes four study areas: Ruyuan, Tongcheng, Jianli, and Hanchuan. Experimental results demonstrate that the proposed architecture achieves superior performance compared with other models, with a mean intersection over union ratio of 89.26% and an accuracy rate of 95.34% (pixel accuracy). The visualization results exhibit exceptional performance in extracting green spaces with clear spatial relationships, such as roadside and community greening while effectively distinguishing between urban green spaces and farmland. Furthermore, we conduct experiments on generalization ability on four different areas. The results show that the generalization ability of SegFormer outperforms other models. Lastly, the experiments verify the impact of different data augmentation multiples on model accuracy, providing valuable insights to guide data augmentation strategies in practical engineering. © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

An Anti-epidemic Delivery Trolley based on YOLOv5 Image Recognition of Room Numbers and QR Codes 4

An Anti-epidemic Delivery Trolley based on YOLOv5 Image Reco...

引用

4th International Conference on computer Vision, Image and Deep Learning, CVIDL 2023

作者： Jiang, Saibiao Fan, Jianan Sun, Zhijin Deng, Baoru School of Electronic Information Engineering Zhuhai College of Science and Technology Zhuhai China School of Computer Engineering Zhuhai College of Science and Technology Zhuhai China

ISBN: (纸本)9798350326444

Since the massive spread of the COVID-19 epidemic in 2020 and the recent spread of influenza, the contactless delivery trolley has met the demands of the government for logistics against epidemics and the needs of logistics companies for deliveries. This paper designs a recognition method based on YOLOv5 (You Only Look Once version 5) for detecting room numbers and QR codes for the needs of anti-epidemic delivery trolleys and implements tasks such as stopping the anti-epidemic trolley at a specified room number and opening different warehouses according to different QR code information. The experiments improved the recognition by filtering the pepper noise through mean filtering. Experimental results show an increase of 2.105 % in the recognition rate of pre-processed QR codes, with an average accuracy rate of 98.9%. The delivery trolley also achieves a stable recognition rate of around 98.81% for room numbers. © 2023 IEEE.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

WLARS: Workload-Aware Recharge Scheduling Mechanism for Improving Surveillance Quality in Wireless Rechargeable Sensor Networks

引用

IEEE Sensors Journal 2023年第11期23卷 12237-12250页

作者： Dande, Bhargavi Chang, Chih-Yung Wu, Shih-Jung Roy, Diptendu Sinha National Taiwan University Department of Electrical Engineering Taipei10617 Taiwan Tamkang University Department of Computer Science and Information Engineering New Taipei25137 Taiwan National Institute of Technology Meghalaya Department of Computer Science and Engineering Shillong793003 India

With radio frequency (RF), the mobile charger (MC) can wirelessly transmit energy to the sensor nodes in the network. The wireless energy transfer enhances the lifetime of sensor nodes in wireless rechargeable sensor networks (WRSNs). Most of the studies improved the efficiency of MC or perpetuated the lifetime of WRSNs. At the same time, none of the studies focused on considering the spatial and temporal surveillance qualities (STSQs) of each sensor recharged. Therefore, this article proposed an efficient energy recharge scheduling for MC, aiming to maximize the STSQ of the given network. Initially, sensor routing load (RL) is considered to partition the network to distribute the charging load of the MCs evenly. Second, the busy level of the MC is considered to send the recharging requests to the MC. Based on the busy level of MC, the sensors adjust the sensing rate frequency to manage their energy. Finally, cooperation between the neighboring MCs is proposed to reduce the waiting time for recharging requested sensors. The simulation results present that the proposed work yields the literature in terms of STSQ, traveling distance, average waiting time (AWT), and energy usage efficiency. © 2001-2012 IEEE.

关键词： computer architecture

来源：评论

学校读者我要写书评

暂无评论

Dynamic Tracking with Fuzzy Rules for Evolutionary Dynamic Constrained Optimization

Dynamic Tracking with Fuzzy Rules for Evolutionary Dynamic C...

引用

2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Bu, Chenyang Liu, Yuxin Zhang, Lizhong Zhu, Tao Luo, Wenjian School of Computer Science and Information Engineering Hefei University of Technology Ministry of Education Key Laboratory of Knowledge Engineering with Big Data Anhui Hefei230027 China School of Computer Science and Information Engineering University of South China Hunan Hengyang421200 China School of Computer Science and Technology Harbin Institute of Technology Guangdong Shenzhen518000 China

ISBN: (纸本)9798350337020

Nature-inspired population-based stochastic search algorithms (SSA) have demonstrated effectiveness in solving many real-world dynamic optimization problems (DOPs), such as dynamic optimal power flow (DOPF) problems. The basic idea of solving DOPs using SSAs is to 'track the moving optima', rather than solving the changed problems from scratch. Its hidden assumption is that the problems are slightly changed in general, and it is expected that the search process can be accelerated by learning from past search experiences. However, the hidden assumption that the current search process can always benefit from previous search experiences might not be true for certain cases. For example, if the current problem is not similar to any of its previous problems, the historical solutions might not be helpful for the current search. To reduce such negative transfer, the following issues are worthy of study: how to choose which historical problems to learn from, and how to determine the degree of learning from historical problems. To solve the above problems, we propose a dynamic processing framework based on fuzzy rules from the perspective of incorporating human knowledge for optima tracking. The experimental results on the DOPF problems show that the SSA with the proposed optima tracking strategy outperforms other comparative algorithms. We open up the code and data of our algorithm44https://***/DMiC-Lab-HFUT/SMDE-Transfer. © 2023 IEEE.

关键词：

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：