检索结果-内蒙古大学图书馆

12th National CCF Conference on Natural Language Processing and Chinese Computing, NLPCC 2023

作者： Qiu, Yunzhi Zhang, Xiaokun Wang, Weiwei Zhang, Tongxuan Xu, Bo Lin, Hongfei School of Computer Science and Technology Dalian University of Technology Dalian China College of Computer and Information Engineering Tianjin Normal University Tianjin China

ISBN: (纸本)9783031446955

Adverse drug reaction (ADR) detection is an essential task in the medical field, as ADRs have a gravely detrimental impact on patients’ health and the healthcare system. Due to a large number of people sharing information on social media platforms, an increasing number of efforts focus on social media data to carry out effective ADR detection. Despite having achieved impressive performance, the existing methods of ADR detection still suffer from three main challenges. Firstly, researchers have consistently ignored the interaction between domain keywords and other words in the sentence. Secondly, social media datasets suffer from the challenges of low annotated data. Thirdly, the issue of sample imbalance is commonly observed in social media datasets. To solve these challenges, we propose the Knowledge Enhanced Shallow and Deep Transformer (KESDT) model for ADR detection. Specifically, to cope with the first issue, we incorporate the domain keywords into the Transformer model through a shallow fusion manner, which enables the model to fully exploit the interactive relationships between domain keywords and other words in the sentence. To overcome the low annotated data, we integrate the synonym sets into the Transformer model through a deep fusion manner, which expands the size of the samples. To mitigate the impact of sample imbalance, we replace the standard cross entropy loss function with the focal loss function for effective model training. We conduct extensive experiments on three public datasets including TwiMed, Twitter, and CADEC. The proposed KESDT outperforms state-of-the-art baselines on F1 values, with relative improvements of 4.87%, 47.83%, and 5.73% respectively, which demonstrates the effectiveness of our proposed KESDT. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

CRAFT: Criticality-Aware Fault-Tolerance Enhancement Techniques for Emerging Memories-Based Deep Neural Networks

引用

IEEE Transactions on computer-Aided Design of Integrated Circuits and Systems 2023年第10期42卷 3289-3300页

作者： Nguyen, Thai-Hoang Imran, Muhammad Choi, Jaehyuk Yang, Joon-Sung Department of Electrical and Computer Engineering Sungkyunkwan University Suwon16419 Korea Republic of Department of Electrical Engineering School of Electrical Engineering and Computer Science National University of Sciences and Technology Islamabad44000 Pakistan Department of Semiconductor Systems Engineering Sungkyunkwan University Suwon16419 Korea Republic of School of Electrical and Electronic Engineering Department of System Semiconductor Engineering Yonsei University Seoul03722 Korea Republic of

Deep neural networks (DNNs) have emerged as the most effective programming paradigm for computer vision and natural language processing applications. With the rapid development of DNNs, efficient hardware architectures for deploying DNN-based applications on edge devices have been extensively studied. Emerging nonvolatile memories (NVMs), with their better scalability, nonvolatility, and good read performance, are found to be promising candidates for deploying DNNs. However, despite the promise, emerging NVMs often suffer from reliability issues, such as stuck-at faults, which decrease the chip yield/memory lifetime and severely impact the accuracy of DNNs. A stuck-at cell can be read but not reprogrammed, thus, stuck-at faults in NVMs may or may not result in errors depending on the data to be stored. By reducing the number of errors caused by stuck-at faults, the reliability of a DNN-based system can be enhanced. This article proposes CRAFT, i.e., criticality-aware fault-tolerance enhancement techniques to enhance the reliability of NVM-based DNNs in the presence of stuck-at faults. A data block remapping technique is used to reduce the impact of stuck-at faults on DNNs accuracy. Additionally, by performing bit-level criticality analysis on various DNNs, the critical-bit positions in network parameters that can significantly impact the accuracy are identified. Based on this analysis, we propose an encoding method which effectively swaps the critical bit positions with that of noncritical bits when more errors (due to stuck-at faults) are present in the critical bits. Experiments of CRAFT architecture with various DNN models indicate that the robustness of a DNN against stuck-at faults can be enhanced by up to 105 times on the CIFAR-10 dataset and up to 29 times on ImageNet dataset with only a minimal amount of storage overhead, i.e., 1.17%. Being orthogonal, CRAFT can be integrated with existing fault-tolerance schemes to further enhance the robustness of DNNs aga

关键词： Fault tolerance

来源：评论

学校读者我要写书评

暂无评论

Beyond Precision and Recall: Measuring Search Engine Consistency Using Rank Stability

引用

IEEE Access 2025年 13卷 92242-92259页

作者： Lakshmi, S. Thinakaran, Rajermani Batri, Krishnan Sowrirajan, R. Murugan, Sivaram Jain University Department of Electronics and Communication Engineering Karnataka Bengaluru India INTI International University Faculty of Data Science and Information Technology Negeri Sembilan Malaysia Jain University School of Computer Science and Engineering Karnataka Bengaluru India Dr. N.G.P. Arts and Science College Department of Mathematics Tamil Nadu Coimbatore India Sivas University of Science and Technology Faculty of Engineering and Natural Sciences Department of Computer Engineering Sivas58000 Turkey

Traditional information retrieval metrics such as precision, recall, and F-measure assess document relevance but fail to capture the stability of search rankings. Search engines frequently update their algorithms, leading to variations in retrieved documents and their rankings across identical queries. This study introduces a novel rank correlation measure that quantifies search result consistency by analyzing rank fluctuations in repeated queries, serving as a proxy for temporally separated searches. The proposed approach distinguishes between overlapping and non-overlapping documents, providing a detailed measure of ranking stability. Empirical evaluations comparing Google and Bing reveal that Google exhibits greater rank fluctuations for non-overlapping documents, whereas both search engines maintain relatively stable rankings for overlapping documents. These findings highlight the importance of rank-based evaluation frameworks in search engine analysis, extending beyond conventional retrieval metrics. By incorporating stability assessments, this research enhances the methodology for evaluating search consistency in dynamic web environments, contributing to process innovation in information retrieval (SDG 9: Industry, Innovation, and Infrastructure) and promoting fair, reliable access to information (SDG 16: Peace, Justice, and Strong Institutions). © 2013 IEEE.

关键词： Online searching

来源：评论

学校读者我要写书评

暂无评论

Investigation of symmetric and asymmetric eye patterns for bell's palsy diagnosis 21

Investigation of symmetric and asymmetric eye patterns for b...

引用

21st OITS International Conference on Information technology, OCIT 2023

作者： Prasad, Thati Ravi Naulegari, Janardhan Visvesvaraya National Institute of Technology Department of Computer Science and Engineering Nagpur India GITAM Deemed-To-be-University GITAM School of Technology Department of Computer Science and Engineering Hyderabad India

ISBN: (纸本)9798350358230

The development of an objective tool for the diagnosis of Bell's palsy (BP) has garnered significant attention among researchers, as it can aid in early treatment. This study aimed to address three research questions in the context of classification of individuals with BP or non-BP conditions. First, whether symmetric and asymmetric parameters in eye features can be reliable indicators for classification of BP and non-BP individuals. The results showed that classifiers use these parameters for classification. Second, To investigate whether the integration of both symmetric and asymmetric patterns of eye features can improve the performance metrics of classification. The findings supported that combination gives better performance. Third, how the incorporation of machine learning classifiers utilizing symmetric and asymmetric eye patterns contributes to the development of an automated classification model for distinguishing BP and non-BP subjects. The findings indicate that the Support Vector Machine(SVM) classifier achieved the highest accuracy of 92% when using the combination of symmetric and asymmetric eye parameters. This supports that the SVM classifier in leveraging the complementary information offered by both patterns to accurately classifying the individuals with BP and non-BP conditions. Overall, this study highlights the potential of symmetric and asymmetric eye patterns as reliable indicators for distinguishing between BP patients and non-BP individuals. © 2023 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Underwater image recovery method considering target polarization characteristics 4

Underwater image recovery method considering target polariza...

引用

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

作者： Wang, Zhe Yu, Haiyang Peng, Jin School of Computer Science and Information Engineering Hefei University of Technology AnHui Hefei China

ISBN: (纸本)9798350326444

In underwater polarization imaging, the estimation of the polarization characteristics of the target is also the key to recover the original image. Most current recovery methods ignore the polarization characteristics of the target and require manual selection of the region for parameter estimation. This not only causes errors in the recovery results, but also has the limitation that a background region must exist. In this paper, we derive the reconstruction model of the original underwater image from the underwater imaging model and the polarization imaging model, considering the polarization characteristics of the target. In this paper, the background light polarization degree is obtained using the filtered median algorithm, and the bright channel principle is used to obtain the background light at infinity. Compared with the traditional method, the method in this paper can reduce the error. Finally, this paper makes improvements to the perfect reflection algorithm and enhances the recovered image. © 2023 IEEE.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Model Simulation of a Space Solar Power System using Disaggregated Apertures of Transmission (DAGATs)

Model Simulation of a Space Solar Power System using Disaggr...

引用

2023 IEEE International Opportunity Research Scholars Symposium, ORSS 2023

作者： Schroeder, Alex Noh, Stephen Park, Paul Brunson, Jada Bhattacharya, Adrija Durgin, Gregory Georgia Institute of Technology School of Electrical and Computer Engineering AtlantaGA United States

ISBN: (纸本)9798350337006

The earth's atmosphere blocks most of the sun's power from reaching the surface, decreasing the energy density for solar power;therefore, a clear alternative is to conduct further research into space solar power (SSP) using satellites and antennas. Larger satellite transmission apertures are vulnerable to suffering catastrophic damage through environmental effects. Thus, a possible solution for SSP is using disaggregated apertures for transmission (DAGATs) which involves having many small satellites in space coordinate together in a volumetric array. These structures would use antennas to transmit the power as a single signal to a surface station, which would provide increased stability and improved cost-effectiveness. This research begins the process of simulating these types of arrays in order to determine their effect on transmission efficiency. A simplified model of a DAGAT system is created, and changes in its elements' positioning are simulated. These changes are the planar spacings, the relative elevations, and the total removal of elements. Although the simulations are highly simplified compared to what a real-world SSP array would look like, they provide patterns showing a decreased directivity with further spacing, but little change in directivity with changes in elevation or with element removal. © 2023 IEEE.

关键词： Earth atmosphere

来源：评论

学校读者我要写书评

暂无评论

Remote sensing image registration based on multi-head self-attention mechanism 8

Remote sensing image registration based on multi-head self-a...

引用

8th International Conference on Intelligent Informatics and Biomedical Sciences, ICIIBMS 2023

作者： Hou, Jianxing Chen, Ying Deng, Xiuhan Ni, Lizheng Shanghai Institute of Technology School of Computer Science & Information Engineering Shanghai201418 China

ISBN: (纸本)9798350369564

Remote sensing image registration is the process of matching and superimposing multiple sets of images. This technology plays an important role in ground object detection, aerial image classification and satellite image fusion, mainly including traditional methods and methods based on deep learning. Among them, the traditional remote sensing image registration algorithm consumes a lot of manpower and runs for too long during registration. Although the remote sensing image registration algorithm based on deep learning reduces labor costs and improves the adaptive learning ability of the model, the registration accuracy and running time of the algorithm still need to be improved. Aiming at the problems existing in the registration algorithm based on deep learning, a registration algorithm combining the multi-head self-attention mechanism and the convolutional triple attention module is proposed. It replaces residual network residual blocks and combines with other pre-trained convolutional layers for feature extraction. The cross-correlation network was designed, and the Pearson cross-correlation algorithm was used to establish the correspondence between remote sensing images. The registration method proposed in this paper is suitable for three types of remote sensing image registration applications: time difference (multi-time phase), different angle of view (multi-viewing angle), and different shooting sensor (multi-modality). Under these three types of registration applications, the proposed algorithm has high registration accuracy and registration efficiency. © 2023 IEEE.

关键词： Image registration

来源：评论

学校读者我要写书评

暂无评论

Enhancing Knowledge-Aware Recommendation with Contrastive Learning 9th

Enhancing Knowledge-Aware Recommendation with Contrastive L...

引用

19th International Conference on Advanced Data Mining and Applications, ADMA 2023

作者： Zhang, Xinyue Gao, Hui School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9783031466601

Knowledge graph serves as a side information, bringing diversity and interpretability to the recommendation. A well-developed recommender system can efficiently capture user and item characteristics, accurately reflecting user preferences. However, supervised signals with graph structure are extraordinarily sparse, and the collaborative and knowledge graphs contain irrelevant edges, exacerbating noise propagation and reducing the robustness of recommendations. To address the above issues, we propose a model for enhancing Knowledge-aware Recommendation with Contrastive Learning (KRCL), including two contrastive learning tasks and three functional modules. Specifically, we construct two views, using TransR and TATEC to optimize knowledge representations from distance and semantic aspects, respectively. After the item-side knowledge is augmented, we remove unreliable interaction edges from collaborative graph to reduce noise propagation. We then perform contrastive learning on the output node representations of different views through graph propagation. To further tap the latent interest of users, we consider users/items that exhibit similar representations as semantic neighbors, treating them as positive pairs in contrastive learning. The structural and semantic contrastive tasks are eventually integrated in a multi-task learning manner to jointly boost the recommendation performance. To validate the effectiveness of our method, we conduct extensive experiments on three benchmark datasets. Experimental results demonstrate that our KRCL significantly outperforms previous state-of-the-art baselines. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Taxonomy for Classification of Cloud Service (Paas) Interoperability 1

引用

World Conference on Information Systems for Business Management, ISBM 2022

作者： Adhoni, Zameer Ahmed Dayanand Lal, N. Department of Computer Science and Engineering Gitam School of Technology Gitam University Bangalore India

ISBN: (数字)9789811974472

ISBN: (纸本)9789811974465

The decreasing cost of hardware has opened the doors to the development of various concepts like big data and huge storage spaces. The development of the cloud is one major development in this string of developments. The economic and other important contract features like the locking time gave rise to the usage of multi-clouds by various companies for the same or similar applications. The extended form of this scenario leads to operation on data on different clouds giving rise to interoperability. The Infrastructure as a service, Platform as a service, and Software as a service in the cloud force us to define taxonomy based on various views. Each view addresses its own customized problem. There is a need to generalize the taxonomy of interoperability. This paper presents a taxonomy for the classification of the cloud for Platform as a Service. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Interoperability

来源：评论

学校读者我要写书评

暂无评论

Detection and Defense of XSS Attack Script Based on Machine Learning

Detection and Defense of XSS Attack Script Based on Machine ...

引用

2023 IEEE International Conference on Sensors, Electronics and computer engineering, ICSECE 2023

作者： Zhao, Huanxi Wang, Yi Xiamen University of Technology School of Computer and Information Engineering Fujian Xiamen361000 China

ISBN: (纸本)9798350313734

With the rapid development of the Internet, many things of people can be done online, and the Internet has been integrated into people's daily life. Web applications widely used in all walks of life are convenient for people to use, but also bring some security problems. Cross-Site Scripting(XSS) vulnerability detection method based on dynamic analysis is an effective XSS vulnerability detection method, which simulates real attack behavior to conduct penetration test on Web applications. However, there are still some problems such as low detection coverage, low detection efficiency and low detection accuracy. Based on this, this paper expounds the research status and background of XSS attack detection, puts forward the necessity of using machine learning algorithm for XSS attack detection, applies SVM algorithm to XSS detection, analyzes and extracts the features of XSS samples, selects radial basis kernel function for nonlinear SVM training, and verifies the generated detection model. It is found that the SVM model in this paper is also superior to the comparison algorithm in accuracy and false alarm rate. © 2023 IEEE.

关键词： Learning algorithms

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：