检索结果-内蒙古大学图书馆

4th IEEE International Conference on Pattern Recognition and Machine Learning, PRML 2023

作者： Zhou, Yan Dai, Zixi Wang, Qingjuan Ren, Lingling School of Computer Beijing Institute of Technology Zhuhai Zhuhai China Kunming University of Science and Technology School of Information Engineering and Automation Kunming China Tibet University School of Information Science and Technology Lhasa China

ISBN: (纸本)9798350324303

In recent years, there has been a growing use of intelligent chatbots in various scenarios. However, most of the current chatbots are limited to simple and mechanical interactions. To address this, by analyzing the formation principle of human emotion, an intelligent chatbot with emotion is studied and designed in this paper. Firstly, open data sets are collected, word segmentation is processed on them, and the feature vectors are extracted by word embedding technology. Then, different neural networks are designed and optimized for two algorithm modules: analyzing emotional tendency of input text, and generating reply text. Finally, the chatbot is achieved. The experimental results show that the accuracy rate, recall rate and F1 score of the algorithm of emotional tendency analyzing are all around 90%, and the reply text quality generated by the chatbot is also relatively ideal. © 2023 IEEE.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

The geoscience knowledge system,ontology and knowledge graph for data-driven discovery:Preface

引用

Geoscience Frontiers 2023年第5期14卷 332-333页

作者： Xiumian Hu Xiaogang Ma Chao Ma Hairong Lv School of Earth Sciences and Engineering Nanjing UniversityPR China Department of Computer Science University of IdahoUSA Institute of Sedimentary Geology Chengdu University of TechnologyPR China Department of Automation Tsinghua UniversityPR China

Earth science data have shown rapid growth since the 21st century with the improvement of experimental instruments and testing *** provides a basis for revealing the evolutionary history of life,climate,palaeogeography and economic deposits by using big ***,it is a major challenge to integrate Earth science data for the complexity of the Earth system,the great number of terminologies in Earth science,the diversity of research methods and proxies,and the diversification of data types.

关键词： knowledge methods. Earth

来源：评论

学校读者我要写书评

暂无评论

Design of Haptic Sensing Snake Robot 7

Design of Haptic Sensing Snake Robot

引用

7th International Conference on Robotics, Control and automation Engineering, RCAE 2024

作者： Zhang, Yang Wu, Hao Li, Peng Bai, Deen Guo, Shuai Tang, Chaoquan China University of Mining and Technology School of Mechanical and Electrical Engineering Xuzhou China Harbin Institute of Technology Shenzhen School of Mechanical Engineering and Automation Shenzhen China Zhengzhou University School of Computer Science and Artificial Intelligence Zhengzhou China

ISBN: (纸本)9798350355642

Snake robots, a category of bionic robots, have garnered significant interest due to their capacity to adapt autonomously to complex unstructured environments. This is achieved through the flexibility and versatility of their bodies and gaits, as well as the coordinated movements of their bodies. However, in unknown or complex environments with obstacles, snake robots are unable to plan their next locomotor actions based solely on the sensory information obtained from the camera they carry. The incorporation of haptics is anticipated to facilitate the snake robot's perception of the interaction between its body and the external environment, thereby enabling more precise and effective guidance for efficient movement. This paper presents the design of a flexible tactile sensor based on the principle of photoconductivity, which has been developed to enable a snake robot to perceive and interact with its environment through tactile sensing. The sensor is intended to enhance the robot's autonomous movement capabilities in complex environments where visual recognition or human observation alone is insufficient. © 2024 IEEE.

关键词： flexible haptic sensor motion control obstacle-assisted motion sinuous motion snake robot

来源：评论

学校读者我要写书评

暂无评论

SIAT: Pedestrian trajectory prediction via social interaction-aware transformer

引用

Complex & Intelligent Systems 2025年第8期11卷 1-14页

作者： Wang, Chengdong Wang, Jianming Gao, Wenbo Guo, Lei School of Computer and Software Engineering Anhui Institute of Information Technology Wuhu China Yunnan Key Laboratory of Computer Technology Applications Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming China

Pedestrian trajectory prediction is crucial for mitigating collision risks in intelligent transportation and surveillance systems. Despite recent advances, accurately capturing and modeling complex social interactions among pedestrians remains a challenge. This paper introduces the Social Interaction-Aware Transformer (SIAT), a novel approach that leverages a Transformer encoder to process pedestrian embedding features and a Graph Convolutional Network (GCN) to construct a social graph for extracting spatial interaction features. The future pedestrian trajectory is predicted using a Transformer decoder that integrates both pedestrian embeddings and social graph features. Extensive experiments on the ETH/UCY and Stanford Drone datasets demonstrate that SIAT significantly outperforms state-of-the-art methods in terms of accuracy and robustness, particularly in densely populated environments. SIAT’s contributions include improved precision through temporal and spatial processing, deep contextual understanding of pedestrian dynamics, and robustness across various settings. The novel model framework establishes a new benchmark for mixed models in trajectory prediction.

关键词： Pedestrian trajectory prediction Social Interaction Transformer GCN

来源：评论

学校读者我要写书评

暂无评论

Digital Educational Resources in Industrial Radiography 7

Digital Educational Resources in Industrial Radiography

引用

7th International Conference on Information Technologies in Engineering Education, Inforino 2024

作者： Kuvshinnikov, Vladimir Kovshov, Evgeny Korchagin, Valery JSC 'Research and Development Institute of Construction Technology - Atomstroy' Research Laboratory of Digital Computer Systems and Automation Moscow Russia Dmitry Mendeleev University of Chemical Technology of Russia Disruptive Engineering School of Chemical Engineering and Machinery Moscow Russia

ISBN: (纸本)9798350371369

Nowadays, the technological level of production is an important factor for all areas of modern economic development and life quality improvement. Neither in the informational, nor in the industrial sphere it is possible to produce in-demand, modern, and therefore high-tech and complex products without the establishment of an appropriate system of quality examination at multiple stages of production. This article deals with the problem of forming a comprehensive educational environment with the use of heterogeneous educational resources in the training of specialists in the radiation type of nondestructive testing. The purpose of the research and development is to address the shortcomings of classroom, remote digital learning, and laboratory practice listed in the article by developing a solution based on virtual reality technologies. The purpose stated lies in the pursuit of intensive solutions to the existing problems. Extensive development paths are not effective in improving the efficiency of training and checking students' knowledge in the context of the rapidly growing demand for nondestructive testing specialists. The concept of a digital twin as an educational resource is considered. The structural approach to the formation of complex educational resources both at the level of organization and at the level of digital product is presented. The structure of interrelationships between data in the formation of a digital twin of a radiographic image is presented. The results of the radiographic image simulation and its quality verification are shown. © 2024 IEEE.

关键词： Nondestructive examination

来源：评论

学校读者我要写书评

暂无评论

Study on Landslide Susceptibility Evaluation Based on Improved Multivariable Weighted Model: A Case Study of Lushan County, China 3

Study on Landslide Susceptibility Evaluation Based on Improv...

引用

3rd IEEE International Conference on Software Engineering and Artificial Intelligence, SEAI 2023

作者： Zhao, Li Duan, Gonghao Wuhan Institute of Technology School of Computer Science and Engineering Wuhan China Automation for Complex Systems Hubei Key Laboratory of Advanced Control and Intelligent Wuhan China

ISBN: (纸本)9798350337952

The evaluation of regional geological hazard susceptibility is of great significance to the prevention and control of geological hazard. In this paper, the "4-20"Lushan earthquake disaster area as the research area, combined with GIS and characteristics of the research area, through correlation analysis, selected 9 influencing factors as the evaluation factors. The study was conducted using a weighted information volume-logistic regression model (WI-LR). The results were classified into five sensitivity levels: very low, low, medium, high and very high. The results show that the landslide prone areas in Lushan County are concentrated in the area below the middle of the county seat, mostly near rivers, faults and areas with peak acceleration greater than 0.4. Among them, WI-LR model (AUC=0.918) > deterministic factor model (CF =0.908) > weighted information quantity model (WI) (AUC=0.896) > information quantity model I (AUC=0.853) > information quantity logistic regression model I-LR (AUC=0.799). It indicates that WI-LR model has high evaluation accuracy. © 2023 IEEE.

关键词： Hazards

来源：评论

学校读者我要写书评

暂无评论

3D Modeling for a Flexible Detector in VR Radiography Simulator 7

3D Modeling for a Flexible Detector in VR Radiography Simula...

引用

7th International Conference on Information Technologies in Engineering Education, Inforino 2024

作者： Kuvshinnikov, Vladimir Kovshov, Evgeny Korchagin, Valery Institute of Construction Technology - Atomstroy Research Laboratory of Digital Computer Systems and Automation JSC 'Research and Development' Moscow Russia Dmitry Mendeleev University of Chemical Technology of Russia Disruptive Engineering School of Chemical Engineering and Machinery Moscow Russia

ISBN: (纸本)9798350371369

Digital simulators based on virtual reality expand the range of educational opportunities and improve the quality of learning in various areas of knowledge through deeper "immersion"in the learning process. Attention to the details of the modeled process plays an important role. One of the tasks of interactive simulators is to provide an opportunity to see the relationship between the user's actions and the obtained qualitative and quantitative parameters and quality attributes of the simulated process result. Considered the issue of curved shape detectors application relevance in the practice of penetrating radiation evaluation in general and in industrial radiography in particular. The prerequisites for the creation of flexible (bendable) form of digital radiographic detectors and the features of two generations of practically implemented devices are analyzed. Development of a flexible detector digital twin is considered, models and algorithms for its realization in a virtual reality environment are proposed, and the main results of computational and laboratory experiments as well as the obtained radiographic images are presented. The main task of virtual reality technologies within the qualified technical specialists training programs is the development of their professional skills and the formation of stable behavior patterns that ensure strict compliance with the rules and requirements of industrial safety, thereby ensuring the high quality of the end product. © 2024 IEEE.

关键词： Virtual reality

来源：评论

学校读者我要写书评

暂无评论

A Novel Deep Learning Approach for Tissue Microstructure Estimation Using Spherical Mean Technique 22

A Novel Deep Learning Approach for Tissue Microstructure Est...

引用

22nd IEEE International Symposium on Biomedical Imaging, ISBI 2025

作者： Pan, Yiang Feng, Yuanjing He, Jianzhong Cui, Jianan Xie, Lei Chen, Zan Zeng, Qingrun Zhang, Jiawei Yu, Yan Kukanov, Konstantin K. Wu, Ye Zhejiang University of Technology Institution of Information Processing and Automation Hangzhou310023 China Polenov Neurosurgery Institute The Branch of Almazov National Medical Research Centre Russia School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210094 China

ISBN: (纸本)9798331520526

Multi-shell diffusion magnetic resonance imaging (dMRI) models connect tissue organization with the observed dMRI signals to infer tissue microstructure. Given the highly non-linear nature of these models, several deep network based approaches have been proposed to improve the efficiency and accuracy of tissue microstructure estimation. However, these approaches rely on dMRI data with dense q-space sampling, which can lead to high computational costs. In this work, we propose a novel deep learning framework, termed DeepSMT, which utilizes the spherical mean technique (SMT) to directly estimate NODDI and DKI microstructure. The network comprises an inter-voxel block and a multi-shell feature block that can capture the feature relationship of the average signal between inter-voxels and integrate features from each shell. We validate the effectiveness of our method using the Human Connectome Project dataset. Overall, the DeepSMT has better performance compared to other methods in estimating tissue microstructure. © 2025 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Multi-hop question answering for SRLGRN augmented by textual relationship modelling 5

Multi-hop question answering for SRLGRN augmented by textual...

引用

5th International Conference on computer Information Science and Artificial Intelligence, CISAI 2022

作者： Zhang, Xuesong Li, Guanjun Zhang, Dawei Lv, Zhao Tao, Jianhua College of Computer Science and Technology Anhui University Hefei230000 China Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing100000 China

ISBN: (纸本)9781510662520

Multi-hop question answering aims to predict answers to questions and generate supporting facts for answers by reasoning over the content of multiple documents. The recently proposed Semantic Role Labeling Graph Reasoning Network (SRLGRN) has achieved excellent performance on multi-hop QA tasks. However, SRLGRN is lacking in modelling the textual relationships, which are import cues for reasoning. To this end, this paper proposes an enhanced SRLGRN multi-hop question answering approach by modelling textual relationships at different granularity (document relationships and sentence relationships). By modelling document relationships, a novel document filter based on document relationship threshold is designed for SRLGRN to dynamically select documents relevant to the question from multiple documents. By modelling sentence relationships, a sentence relationship-aware answer type prediction module is added to SRLGRN, which models sentences in documents as sentence graphs and then uses graph convolution network to predict answer type. The obtained answer type further guide the answer reasoning module of SRLGRN to obtain question answer with supporting facts. The experimental results show that the proposed scheme outperforms SRLGRN in terms of answer prediction and supporting fact prediction, with a 2% improvement in answer F1 metrics and a 3.1% improvement in joint F1 performance. © 2023 SPIE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

A DDPG Algorithm Based Reinforcement Learning Controller for Three-Phase DC-AC Inverters

A DDPG Algorithm Based Reinforcement Learning Controller for...

引用

2023 International Conference on Power Energy Systems and Applications, ICoPESA 2023

作者： Ye, Jian Mei, Sen Guo, Huanyu Hu, Yingjie Zhang, Xinan Harbin Institute of Technology Shenzhen School of Mechanical Engineering and Automation Shenzhen China University of Western Australia School of Electrical Electronic and Computer Engineering Perth Australia

ISBN: (纸本)9798350345605

This paper proposes a DC-AC inverter controller based on reinforcement learning (RL) algorithm. Compared with the traditional PID control method, the structure of the RL algorithm based controller is simpler. The deep deterministic policy gradient (DDPG) algorithm in RL algorithm is used to realize model-free control of inverters, so that the control algorithm has adaptive ability to different types of DC-AC inverters. It can avoid the dependence of the control strategy on the system model. Through simulation, the control strategy of three-phase two-level DC-AC inverter based on DDPG algorithm is compared with the traditional PID control. The simulation results show that the total harmonic distortion (THD) is reduced by 12% and the current tracking performance is improved by at least 75%. © 2023 IEEE.

关键词： Reinforcement learning

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：