检索结果-内蒙古大学图书馆

Knowledge Graph and Knowledge Reasoning:A Systematic Review

Journal of Electronic science and technology 2022年第2期20卷 159-186页

作者： Ling Tian Xue Zhou Yan-Ping Wu Wang-Tao Zhou Jin-Hao Zhang Tian-Shu Zhang Kashi Institute of Electronics and Information Industry Kashi 844099 Shenzhen Institute of Information Technology Shenzhen 518172 the School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengdu 610054 the School of Information and Software Engineering University of Electronic Science and Technology of ChinaChengdu 610054

The knowledge graph(KG) that represents structural relations among entities has become an increasingly important research field for knowledge-driven artificial intelligence. In this survey, a comprehensive review of KG and KG reasoning is provided. It introduces an overview of KGs, including representation, storage, and essential technologies. Specifically, it summarizes several types of knowledge reasoning approaches, including logic rules-based, representation-based, and neural network-based methods. Moreover, this paper analyzes the representation methods of knowledge hypergraphs. To effectively model hyper-relational data and improve the performance of knowledge reasoning, a three-layer knowledge hypergraph model is proposed. Finally, it analyzes the advantages of three-layer knowledge hypergraphs through reasoning and update algorithms which could facilitate future research.

关键词： methods. representation reasoning

来源：评论

学校读者我要写书评

暂无评论

Vulnerability Detection of Ethereum Smart Contract Based on SolBERT-BiGRU-Attention Hybrid Neural Model

引用

computer Modeling in engineering & sciences 2023年第10期137卷 903-922页

作者： Guangxia Xu Lei Liu Jingnan Dong Cyberspace Institute of Advanced Technology Guangzhou UniversityGuangzhou510006China School of Software Engineering Chongqing University of Posts and TelecommunicationsChongqing400065China School of Computer Science and Technology Chongqing University of Posts and TelecommunicationsChongqing400065China

In recent years,with the great success of pre-trained language models,the pre-trained BERT model has been gradually applied to the field of source code ***,the time cost of training a language model from zero is very high,and how to transfer the pre-trained language model to the field of smart contract vulnerability detection is a hot research direction at *** this paper,we propose a hybrid model to detect common vulnerabilities in smart contracts based on a lightweight pre-trained languagemodel BERT and connected to a bidirectional gate recurrent *** downstream neural network adopts the bidirectional gate recurrent unit neural network model with a hierarchical attention mechanism to mine more semantic features contained in the source code of smart contracts by using their *** experiments show that our proposed hybrid neural network model SolBERT-BiGRU-Attention is fitted by a large number of data samples with smart contract vulnerabilities,and it is found that compared with the existing methods,the accuracy of our model can reach 93.85%,and the Micro-F1 Score is 94.02%.

关键词： Smart contract pre-trained language model deep learning recurrent neural network blockchain security

来源：评论

学校读者我要写书评

暂无评论

Degradation of o-dichlorobenzene by DBD-NTP co-modified titanium gel catalyst

引用

Journal of Environmental sciences 2024年第9期143卷 71-84页

作者： Wenbo Zhang Yi Xing Wei Su Jiaqing Wang Haoqi Jia Yongkang Cui Jing Chen Hui Zhang School of Energy and Environmental Engineering University of Science and Technology BeijingBeijing 100083China Beijing Key Laboratory of Resource-oriented Treatment of Industrial Pollutants University of Science and Technology BeijingBeijing 100083China Guangdong Province Engineering Laboratory for Air Pollution Control Guangzhou 510530China College of Environmental and Resource Sciences Shanxi UniversityTaiyuan 030006China Sinosteel Maanshan Mine Research Institute Co.Ltd. Anhui 243071China

In order to study the degradation process of dioxins in industrial flue gas,the decomposition of o-dichlorobenzene(o-DCB)in a DBD plasma catalytic reactor was *** results showed that an NTP-catalyzed system,especially using the Cu Mn Ti Oxcatalyst,had better o-DCB degradation performance compared to plasma *** combination of the Cu Mn Ti Oxcatalyst with NTP can achieve a degradation efficiency of up to 97.2%for o-DCB;the selectivity of CO and CO_(2)and the carbon balance were 40%,45%,and 85%,*** dielectric constant and electrical property results indicated that the surface discharge capacity of the catalysts played a major role in the degradation of o-DCB,and a higher dielectric constant could suppress the plasma expansion and enhance the duration of the plasma discharge per discharge *** to the O1s XPS and O_(2)-TPD results,the conversion of CO to CO_(2)follows the M-v-K mechanism;thus,the active species on the catalyst surface play an important ***,the Cu Mn Ti Oxand NTP mixed system exhibited excellent stability,which is probably because Cu doping improved the lifetime of the *** work can provide an experimental and theoretical basis for research in the degradation of o-DCB by plasma catalyst systems.

关键词： o-dichlorobenzene(o-DCB) Non-thermal plasma catalysis Ozone concentration

来源：评论

学校读者我要写书评

暂无评论

Dilated Convolutional Transformer for High-Quality Image Deraining

Dilated Convolutional Transformer for High-Quality Image Der...

引用

2023 IEEE/CVF Conference on computer Vision and Pattern Recognition Workshops, CVPRW 2023

作者： Li, Yufeng Lu, Jiyang Chen, Hongming Wu, Xianhao Chen, Xiang Shenyang Aerospace University College of Electronic and Information Engineering China Nanjing University of Science and Technology School of Computer Science and Engineering China

ISBN: (纸本)9798350302493

Convolutional neural networks (CNNs) and Transformers have achieved significant success in image signal processing. However, little effort has been made to effectively combine the properties of these two architectures to satisfy image deraining. In this paper, we propose an effective de-raining method, dilated convolutional Transformer (DCT), which can enlarge the receptive fields of the network to aggregate global information. The fundamental building unit of our approach is the dilformer block containing multi-dilconv sparse attention (MDSA) and multi-dilconv feed-forward network (MDFN). The MDSA calculates the multi-scale query to generate accurate similarity map so that rich multi-scale information can be better utilized for the high-quality image reconstruction. In addition, we adopt ReLU to replace the original softmax to enforce sparsity in the Transformer for better feature aggregation. The MDFN is further established to better integrate the rain information of different scales in the feature transformation. Extensive experiments on the benchmarks show the favorable performance against state-of-the-art approaches. © 2023 IEEE.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Towards Robust Temporal Activity Localization Learning with Noisy Labels 30

Towards Robust Temporal Activity Localization Learning with ...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Liu, Daizong Qu, Xiaoye Fang, Xiang Dong, Jianfeng Zhou, Pan Nan, Guoshun Tang, Keke Fang, Wanlong Cheng, Yu Wangxuan Institute of Computer Technology Peking University China School of Cyber Science and Engineering Huazhong University of Science of Technology China Interdisciplinary Graduate Programme Nanyang Technological University Singapore College of Computer Science and Technology Zhejiang Gongshang University China School of Cyberspace Security Beijing University of Posts and Telecommunications China Cyberspace Institute of Advanced Technology Guangzhou University China Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong

ISBN: (纸本)9782493814104

This paper addresses the task of temporal activity localization (TAL). Although recent works have made significant progress in TAL research, almost all of them implicitly assume that the dense frame-level correspondences in each video-query pair are correctly annotated. However, in reality, such an assumption is extremely expensive and even impossible to satisfy due to subjective labeling. To alleviate this issue, in this paper, we explore a new TAL setting termed Noisy Temporal activity localization (NTAL), where a TAL model should be robust to the mixed training data with noisy moment boundaries. Inspired by the memorization effect of neural networks, we propose a novel method called Co-Teaching Regularizer (CTR) for NTAL. Specifically, we first learn a Gaussian Mixture Model to divide the mixed training data into preliminary clean and noisy subsets. Subsequently, we refine the labels of the two subsets by an adaptive prediction function so that their true positive and false positive samples could be identified. To avoid single model being prone to its mistakes learned by the mixed data, we adopt a co-teaching paradigm, which utilizes two models sharing the same framework to teach each other for robust learning. A curriculum strategy is further introduced to gradually learn the moment confidence from easy to hard. Experiments on three datasets demonstrate that our CTR is significantly more robust to the noisy training data compared to the existing methods. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Gaussian distribution

来源：评论

学校读者我要写书评

暂无评论

Towards Unified Deep Image Deraining: A Survey and a New Benchmark

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年第7期47卷 5414-5433页

作者： Xiang Chen Jinshan Pan Jiangxin Dong Jinhui Tang School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

Recent years have witnessed significant advances in image deraining due to the progress of effective image priors and deep learning models. As each deraining approach has individual settings (e.g., training and test datasets, evaluation criteria), how to fairly evaluate existing approaches comprehensively is not a trivial task. Although existing surveys aim to thoroughly review image deraining approaches, few of them focus on unifying evaluation settings to examine the deraining capability and practicality evaluation. In this paper, we provide a comprehensive review of existing image deraining methods and provide a unified evaluation setting to evaluate their performance. Furthermore, we construct a new high-quality benchmark named HQ-RAIN to conduct extensive evaluations, consisting of 5,000 paired high-resolution synthetic images with high harmony and realism. We also discuss existing challenges and highlight several future research opportunities worth exploring. To facilitate the reproduction and tracking of the latest deraining technologies for general users, we build an online platform to provide the off-the-shelf toolkit, involving the large-scale performance evaluation.

关键词： Rain Benchmark testing Training Atmospheric modeling Surveys Mathematical models Reviews Performance evaluation Deep learning Synthetic data

来源：评论

学校读者我要写书评

暂无评论

Short-term passenger flow prediction strategy of urban rail transit based on ridge regression and LSTM

Short-term passenger flow prediction strategy of urban rail ...

引用

2023 IEEE International Conference on Control, Electronics and computer technology, ICCECT 2023

作者： Liu, Hongkun Ren, Jinshan Wang, Zixuan University of Jinan School of Information Science and Engineering Jinan250022 China Jiangsu University of Science and Technology School of Computer Science Zhenjiang212000 China Shandong University of Science and Technology Department of Electrical Engineering and Information Technology Jinan250031 China

ISBN: (纸本)9798350399028

In order to reduce the interference of noise in the short-term passenger flow prediction of urban rail transit, an efficient and accurate traffic flow prediction scheme has become the most concerned issue for traffic participants. This paper focuses on the application of ridge regression and LSTM neural network in the short-term passenger flow of urban rail transit. This paper firstly performs data cleaning, missing processing and normalization. Then, we choose LSTM as the core and use the ridge regression-assisted model to predict the short-term passenger flow of urban rail transit. Ridge regression is used for prediction when there is less data, and LSTM with high fitting degree is used for medium and long-term prediction. The model has a high fitting effect on the randomly divided test set and real data. © 2023 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Sentiment analysis in social internet of things using contextual representations and dilated convolution neural network

引用

Neural Computing and Applications 2024年第20期36卷 12357-12370页

作者： Abid, Fazeel Rasheed, Jawad Hamdi, Mohammed Alshahrani, Hani Al Reshan, Mana Saleh Shaikh, Asadullah Department of Information Systems Dr. Hassan Murad School of Management University of Management and Technology Lahore Pakistan Department of Computer Engineering Istanbul Sabahattin Zaim University Istanbul34303 Turkey Department of Software Engineering Istanbul Nisantasi University Istanbul34394 Turkey Department Computer Science College of Computer Science and Information Systems Najran University Najran61441 Saudi Arabia Department Information Systems College of Computer Science and Information Systems Najran University Najran61441 Saudi Arabia

The methodologies based on neural networks are substantial to accomplish sentiment analysis in the Social Internet of Things (SIoT). With social media sentiment analysis, significant insights can produce efficient and intelligent applications. Neural networks such as recurrent neural networks (RNNs) and convolution neural networks (CNNs) have been considered widely in many text classification tasks. However, RNNs are computationally expensive and require complex training to capture contextual information and long-term dependencies. Similarly, traditional CNNs must stack multiple convolutional layers, requiring massive computations and additional parameters. To address these problems, this work initialized the novel architecture, in which contextual representations (CRs) based on the textual framework are proposed at the initial step. In CRs, state-of-the-art word representation models, such as GloVe (global vectors) and FastText (subword information), collectively produce word representations upon the input sequence using a weight mechanism. Secondly, a unique way is introduced: a three-parallel layered dilated convolutional network with global mean pooling. The experimental results show that the proposed methods when compared with baseline methods, the dilation in CNNs following CRs significantly increases the accuracy from 72.45 to 98.98% and reduces computational resources. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Covert Terahertz Communication for UAV-Aided Wireless Relay Systems

Covert Terahertz Communication for UAV-Aided Wireless Relay ...

引用

2023 International Conference on Networking and Network Applications, NaNA 2023

作者： Pi, Xinzhe Yang, Bin Jiang, Xiaohong School of Systems Information Science Future University Hakodate Hakodate Japan School of Computer and Information Engineering Chuzhou University Anhui China

ISBN: (纸本)9798350327380

Unmanned aerial vehicle (UAV)-aided Terahertz (THz) relay systems can provide high-speed wireless line-of-sight (LoS) communications at THz bands, which are envisioned as a key component of 6G and beyond. However, security is still a significant challenge when adversaries reside in the THz signal beam. This paper investigates the covert THz communications in a UAV-aided two-hop relay system consisting of a transmitter, a UAV relay, a receiver, and two UAV wardens detecting the existence of transmissions in both hops. To enhance security, we consider a friendly jamming scheme that the UAV relay and receiver also serve as jammers to interfere with the two UAV wardens. We then derive the closed-from expressions for the optimal thresholds of wardens and the minimum cascaded detection error probability (DEP). Additionally, we derive the cascaded outage probability and average covert capacity of the covert THz communications. Numerical results are presented to demonstrate the impact of crucial system parameters on the covert THz performances. © 2023 IEEE.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Machine Learning-Based Cellular Automata Model and Frame Work for Tsunami Wave Propagation in Indian Coastal Region

Machine Learning-Based Cellular Automata Model and Frame Wor...

引用

2024 International Conference on Advances in Modern Age Technologies for Health and engineering science, AMATHE 2024

作者： Syed Mohamed, E. Chithirai Pon Selvan, M. B. S. Abdur Rahman Crescent Institute Of Science And Technology Computer Science And Engineering Chennai India School Of Science And Engineering Curtin University Dubai United Arab Emirates

ISBN: (数字)9798350371567

ISBN: (纸本)9798350371567

Simulating the spread of tsunami waves quantitatively holds significant importance in devising swift risk management strategies and implementing efficient solutions for predicting the earliest arrival time. One effective method for dynamically simulating the spatiotemporal evolution of complex systems is through the use of cellular automata (CA), making it a suitable choice for modeling the intricate process of tsunami wave propagation. Nevertheless, the phenomenon of tsunami wave propagation is intricately linked with numerous interdependent factors, rendering it too intricate for analysis using traditional approaches. In answer to this challenge, we present a brand-new method for simulating the propagation of tsunami waves called LSSVM-CA. In this novel method, we integrate a three-dimensional tsunami wave CA framework with least squares support vector machines (LSSVM). We can now take into account and examine how neighboring waves affect the propagation patterns thanks to this integration. Using a dataset derived from historical data, LSSVM is trained to create sophisticated state transformation rules for tsunami wave propagation. We compared the LSSVM-CA-simulated tsunami wave spread areas with the real observed tsunami wave spread areas in order to evaluate the efficacy of the model. The outcomes unequivocally show how well LSSVM-CA simulates tsunami wave propagation accurately calculates a probability of the waves spreading. © 2024 IEEE.

关键词： Wave propagation

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：