检索结果-内蒙古大学图书馆

5th International Conference on Artificial Intelligence and Pattern Recognition, AIPR 2022

作者： Guan, Wei Huang, Mingrui Ma, Li Xi'an University of Posts and Telecommunications Shaanxi Provincial Key Laboratory of Network Data Analysis and Intelligent Processing China

ISBN: (纸本)9781450396899

With the continuous progress of neural network technology in different fields, people have put forward higher requirements and prospects for the research of deep neural network technology in the field of artificial intelligence, and language intelligence is the core problem of artificial intelligence. As an important research content in the field of natural language processing, dialogue generation has been widely concerned by the academic community for a long time, and this research is also widely used in people's daily lives, such as medical Q&A, e-commerce shopping, emotional response, etc. There are still many problems in the existing model for dialogue generation research, such as easy to produce universal replies ("ok", "emm", etc.), the emotional information of the reply statement is not strong enough, and the response is not related to the subject. For this problem, we propose to use an emotional dialogue generation model based on a generative adversarial network to generate statements with emotional responses, and our network has a generator and multi-level discriminator. In terms of discriminators, multi-level discourse discriminators are introduced to guide the generation of replies and the strengthening of emotional information, to discriminate emotions at the individual word level and sentence level, to solve the problem that the same text may be different in different semantic environments and the emotional information of some words is not clear in some cases, in order to achieve maximum accuracy the emotional discrimination results are fed back to the generator and guide the generation of generator statements. In addition, compared with the traditional generative adversarial network, the task processing ability for large pieces of text is improved, and the model has been shown to produce emotional reply statements with better baseline levels than in the past. © 2022 ACM.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

CA-WGE: A two-view graph neural network-based knowledge graph completion approach combining common sense perception 6

CA-WGE: A two-view graph neural network-based knowledge grap...

引用

6th International Conference on Artificial Intelligence and Pattern Recognition, AIPR 2023

作者： Guan, Wei Lian, Xiaoru Ma, Li Xi'an University of Posts and Telecommunications Shaanxi Provincial Key Laboratory of Network Data Analysis and Intelligent Processing China

ISBN: (纸本)9798400707674

The knowledge graph completion algorithm can make the knowledge graph more complete and is currently a research hotspot in the field of artificial intelligence. The knowledge graph completion model is mainly defined in three aspects, the way of negative example generation, the design of scoring function and the design of loss function. The previous knowledge graph completion models only rely on factual view data to predict the missing links between entities and ignore the valuable common sense knowledge, and there is invalid negative sampling in knowledge graph embedding techniques;on the other hand, the existing graph neural network-based knowledge graph embedding models mainly consider capturing the graph structure around entities, and the relational representation is only used to update the entity embedding, which may miss the potentially useful information about the relational structure of potentially useful information. To address the above challenges, this paper proposes a two-view graph neural network-based knowledge graph completion model combined with common sense awareness. Common knowledge is first automatically extracted from fact triples with entity concepts to facilitate high-quality negative sampling, and then positive and weighted negative triples are fed into the two-view graph neural network-based knowledge graph embedding model to capture entity- and relationship-centric graph structures and learn vector representations of entities and relationships, and then the learned entity and relationship representations are fed into a weighted score function to return the final the final score. Extensive experimental and ablation studies on four datasets, FB15K, FB15K237, NELL995, and DBpedia-242, show that the model achieves better performance compared to the state-of-the-art models. © 2023 ACM.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Dialogue Generation for Heterogeneous Graph Neural on Networks Incorporating External Knowledge 6

Dialogue Generation for Heterogeneous Graph Neural on Networ...

引用

6th International Conference on Artificial Intelligence and Pattern Recognition, AIPR 2023

作者： Guan, Wei Guo, Shuaiqian Ma, Li Xi'an University of Posts and Telecommunications Shaanxi Provincial Key Laboratory of Network Data Analysis and Intelligent Processing China

ISBN: (纸本)9798400707674

In recent years, with the improvement of hardware computing power and the rapid development of deep learning theory, the field of dialogue generation has also entered the era of deep learning. However, because it is difficult to generate content-rich and recovery statements by learning semantics only from dialogue corpus, many models are prone to generate responses that are weakly scalable to the topic of the input dialogue or generic responses that lack knowledge information. To address the above issues, a Heterogeneous Graph Neural Network Incorporating External Knowledge Dialogue Generation Model (HEDG) is proposed. First, we build up the external knowledge into a knowledge graph, and find the relevant external knowledge in the knowledge graph according to the entities in the input conversation according to certain rules. This external knowledge and the historical conversation context are then used as inputs to the heterogeneous graph encoder, and the conversation content and the corresponding external knowledge are represented by a Heterogeneous Graph Neural Network (HGNN). Finally, a decoder with Transformer structure as the main body is used to generate a response that is relevant to the conversation context with some knowledge extension by taking the encoded heterogeneous graphical representation as input. The experimental results show that the model has higher relevance and diversity compared with existing comparison models, and can effectively integrate external knowledge. © 2023 ACM.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Automatic ocular artifact removal from EEG data using a hybrid CAE-RLS approach

引用

The Journal of China Universities of Posts and Telecommunications 2020年第1期27卷 81-91页

作者： Wang Zhongmin Tian Meng Liang Chen Song Hui School of Computer Science and Technology Xi'an University of Posts and TelecommunicationsXi'an 710121China Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'an University of Posts and TelecommunicationsXi'an 710121China

Traditional methods for removing ocular artifacts(OAs) from electroencephalography(EEG) signals often involve a large number of EEG electrodes or require electrooculogram(EOG) as the reference, these constraints make subjects uncomfortable during the acquisition process and increase the complexity of brain-computer interfaces(BCI). To address these limitations, a method combining a convolutional autoencoder(CAE) and a recursive least squares(RLS) adaptive filter is proposed. The proposed method consists of offline and online stages. In the offline stage, the peak and local mean of the four-channel EOG signals are automatically extracted to obtain the CAE model. Once the model is trained, the EOG channels are no longer needed. In the online stage, by using the CAE model to identify the OAs from a single-channel raw EEG signal, the identified OAs and the given raw EEG signal are used as the reference and input for an RLS adaptive filter. Experiments show that the root mean square error(RMSE) of the CAE-RLS algorithm and independent component analysis(ICA) are 1.253 3 and 1.254 6 respectively, and the power spectral density(PSD) curve for the CAE-RLS is similar to the original EEG signal. These experimental results indicate that by using only a couple of EEG channels, the proposed method can effectively remove OAs without parallel EOG records and accurately reconstruct the EEG signal. In addition, the processing time of the CAE-RLS is shorter than that of ICA, so the CAE-RLS algorithm is very suitable for BCI system.

关键词： electroencephalography(EEG) electrooculogram(EOG) ocular artifacts(OAs) recursive least squares(RLS) convolutional autoencoder(CAE)

来源：评论

学校读者我要写书评

暂无评论

Research on Improved MobileViT image Tamper Localization Model

引用

Computers, Materials & Continua 2024年第8期80卷 3173-3192页

作者： Jingtao Sun Fengling Zhang Huanqi Liu Wenyan Hou School of Computer Science and Technology Xi’an University of Posts and TelecommunicationsXi’an710121China Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi’an University of Posts and TelecommunicationsXi’an710121China

As image manipulation technology advances rapidly,the malicious use of image tampering has alarmingly escalated,posing a significant threat to social *** the realm of image tampering localization,accurately localizing limited samples,multiple types,and various sizes of regions remains a multitude of *** issues impede the model’s universality and generalization capability and detrimentally affect its *** tackle these issues,we propose FL-MobileViT-an improved MobileViT model devised for image tampering *** proposed model utilizes a dual-stream architecture that independently processes the RGB and noise domain,and captures richer traces of tampering through dual-stream ***,the model incorporating the Focused Linear Attention mechanism within the lightweight network(MobileViT).This substitution significantly diminishes computational complexity and resolves homogeneity problems associated with traditional Transformer attention mechanisms,enhancing feature extraction diversity and improving the model’s localization *** comprehensively fuse the generated results from both feature extractors,we introduce the ASPP architecture for multi-scale feature *** facilitates a more precise localization of tampered regions of various ***,to bolster the model’s generalization ability,we adopt a contrastive learning method and devise a joint optimization training strategy that leverages fused features and captures the disparities in feature distribution in tampered *** strategy enables the learning of contrastive loss at various stages of the feature extractor and employs it as an additional constraint condition in conjunction with cross-entropy *** a result,overfitting issues are effectively alleviated,and the differentiation between tampered and untampered regions is *** evaluations on five benchmark datasets(IMD-20,CASIA,NIST-16,Columbia and Coverage)validat

关键词： image tampering localization focused linear attention mechanism MobileViT contrastive loss

来源：评论

学校读者我要写书评

暂无评论

Rotating Stomata Measurement Based on Anchor-Free Object Detection and Stomata Conductance Calculation

引用

Plant Phenomics 2023年第4期5卷 767-778页

作者： Fan Zhang Bo Wang Fuhao Lu Xinhong Zhang Huaihe Hospital of Henan University Kaifeng 475004China Henan Key Laboratory of Big Data Analysis and Processing Henan UniversityKaifeng 475004China State Key Laboratory of Crop Stress Adaptation and Improvement Henan UniversityKaifeng 475004China School of Software Henan UniversityKaifeng 475004China

Stomata play an essential role in regulating water and carbon dioxide levels in plant leaves,which is important for *** deep learning-based plant stomata detection methods are based on horizontal *** detection anchor boxes of deep learning model are horizontal,while the angle of stomata is randomized,so it is not possible to calculate stomata traits directly from the detection anchor *** processing of image(e.g.,rotating image)is required before detecting stomata and calculating stomata *** paper proposes a novel approach,named DeepRSD(deep learning-based rotating stomata detection),for detecting rotating stomata and calculating stomata basic traits at the same ***,the stomata conductance loss function is introduced in the DeepRSD model training,which improves the efficiency of stomata detection and conductance *** experimental results demonstrate that the DeepRSD model reaches 94.3%recognition accuracy for stomata of maize *** proposed method can help researchers conduct large-scale studies on stomata morphology,structure,and stomata conductance models.

关键词： calculation. Deep rotating

来源：评论

学校读者我要写书评

暂无评论

Jointly Part-of-Speech Tagging and Semantic Role Labeling Using Auxiliary Deep Neural Network Model

引用

Computers, Materials & Continua 2020年第10期65卷 529-541页

作者： Yatian Shen Yubo Mai Xiajiong Shen Wenke Ding Mengjiao Guo School of Computer Science and Engineering Southeast UniversityNanjing210000China Henan Key Laboratory of Big Data Analysis and Processing Henan UniversityKaifeng475000China Swinburne Data Science Research Institute Swinburne University of TechnologyVictoria3122Australia

Previous studies have shown that there is potential semantic dependency between part-of-speech and semantic *** the same time,the predicate-argument structure in a sentence is important information for semantic role labeling *** this work,we introduce the auxiliary deep neural network model,which models semantic dependency between part-of-speech and semantic roles and incorporates the information of predicate-argument into semantic role *** on the framework of joint learning,part-of-speech tagging is used as an auxiliary task to improve the result of the semantic role *** addition,we introduce the argument recognition layer in the training process of the main task-semantic role labeling,so the argument-related structural information selected by the predicate through the attention mechanism is used to assist the main *** the model makes full use of the semantic dependency between part-of-speech and semantic roles and the structural information of predicate-argument,our model achieved the F1 value of 89.0%on the WSJ test set of CoNLL2005,which is superior to existing state-of-the-art model about 0.8%.

关键词： Part-of-speech tagging semantic role labeling multi-task learning

来源：评论

学校读者我要写书评

暂无评论

Multi-dimensional graph linear canonical transform

引用

Digital Signal processing: A Review Journal 2025年 165卷

作者： Li, Na Zhang, Zhichao School of Mathematics and Statistics Center for Applied Mathematics of Jiangsu Province Jiangsu International Joint Laboratory on System Modeling and Data Analysis Nanjing University of Information Science and Technology Nanjing210044 China Key Laboratory of System Control and Information Processing Ministry of Education Shanghai200240 China Key Laboratory of Computational Science and Application of Hainan Province Haikou571158 China

Multi-dimensional (M-D) graph signals represent fundamental data structures in various practical applications, including digital imaging systems, sensor network telemetry, and meteorological observation records. The development of efficient transformation methodologies for processing such M-D graph signals within the linear canonical transform domain continues to present significant challenges. This paper introduces two novel transforms: the two-dimensional graph linear canonical transform based on central discrete dilated Hermite functions (2-D CDDHFs-GLCT) and the two-dimensional graph linear canonical transform based on chirp multiplication-chirp convolution-chirp multiplication decomposition (2-D CM-CC-CM-GLCT). These foundational constructs are extended to the multi-dimensional domain, resulting in M-D CDDHFs-GLCT and M-D CM-CC-CM-GLCT. In terms of computational complexity, additivity and reversibility, M-D CDDHFs-GLCT and M-D CM-CC-CM-GLCT are compared. Theoretical analyses reveal that the M-D CM-CC-CM-GLCT algorithm achieves substantial computational complexity reduction compared to its CDDHFs-based counterpart. Numerical simulations further establish that while maintaining comparable additivity performance, the M-D CM-CC-CM-GLCT exhibits enhanced reversibility characteristics. To validate practical applicability, the proposed M-D GLCT framework is implemented in data compression scenarios. Experimental results demonstrating superior algorithmic efficiency and implementation effectiveness compared to multi-dimensional graph fractional Fourier transform. © 2025 Elsevier Inc.

关键词： Linear transformations

来源：评论

学校读者我要写书评

暂无评论

Knowledge graph embedding via multiplicative interaction 18

Knowledge graph embedding via multiplicative interaction

引用

2nd International Conference on Innovation in Artificial Intelligence, ICIAI 2018

作者： Huang, Zichao Li, Bo Yin, Jian School of Data and Computer Science Sun Yat-sen University Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou510006 China

ISBN: (纸本)9781450363457

Knowledge graphs are playing a crucial role in many machine learning applications. Since most of the knowledge graphs are far from complete, many knowledge graph completion models have been proposed. TransE and its extended models all model knowledge graphs with additive interaction. DistMult demonstrates that multiplicative interaction is more effective for modeling knowledge graphs. However, DistMult performs poorly on 1-to-N, N-to-1 and N-to-N relations. Besides, it does not consider the case that an entity could be a head entity or a tail entity which should be modeled separately. We propose a more fine-grained knowledge graph embedding model called MultE, which models knowledge graphs with multiplicative interaction. In MultE, an entity would have one representation when serving as head entity and another representation when serving as tail entity. For improving performance on 1-to-N, N-to-1 and N-to-N relations, MultE considers all valid tail(or head) entities during training and treats the prediction task as a classification problem. Experiment results show that MultE obtains consistent improvements compared with DistMult and achieves state-of-theart performance. © 2018 Association for Computing Machinery.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Deduplication with blockchain for secure cloud storage 6th

Deduplication with blockchain for secure cloud storage

引用

6th CCF Academic Conference on Big data, CCF Big data 2018

作者： Li, Jingyi Wu, Jigang Chen, Long Li, Jiaxing Guangdong University of Technology Guangzhou China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou510006 China

ISBN: (纸本)9789811329210

Secure deduplication techniques have been wildly used in cloud storage to save both disk space and network bandwidth. However, traditional schemes store only one copy in cloud which raises problems with data reliability. Meanwhile, the existence of any third party, which is adopted by many deduplication schemes, will bring the risk of being single-pointed. In order to solve the above problem, this paper proposes a deduplication scheme which distributes files to multi servers and records storage informations on blockchain. We present smart contract based protocols without the participants of the central authorities to provide secure deduplication. With safety analysis, we demonstrate that our deduplication scheme is secure and reliable under the definitions specified in the proposed security model with acceptable performance through simulation experiments. © Springer Nature Singapore Pte Ltd. 2018.

关键词： Blockchain

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：