检索结果-内蒙古大学图书馆

作者： Zhenyang Ren Fucheng Wan Hongzhi Yu Tiantian Wu Key Laboratory of China's Ethnic Languages and Information Technology of Ministry of Education Northwest Minzu University Key Laboratory of China's Ethnic Languages and Intelligent Processing of Gansu Province Northwest Minzu University

At present,the communication and preservation of project construction documents in most construction enterprises are still using traditional paper method,so the utilization of information is very *** cannot guarantee the completeness,accuracy and systematic of architectural engineering *** the same time,the environment for information management of construction documents is so complicated,which also increases the difficulty of document classification in the field of *** paper first constructs a corpus in the field of construction,then pre-treats and constructs word vector of documents to be ***,finishing the task of classification with text classification model of Bi-LSTM combined with attention.

关键词： Attention Deep Learning Document Classification Construction

来源：评论

学校读者我要写书评

暂无评论

Direct observation of layer-stacking and oriented wrinkles in multilayer hexagonal boron nitride

arXiv

引用

arXiv 2021年

作者： Chen, Lingxiu Elibol, Kenan Cai, Haifang Jiang, Chengxin Shi, Wenhao Chen, Chen Wang, Hui Shan Wang, Xiujun Mu, Xiaojing Li, Chen Watanabe, Kenji Taniguchi, Takashi Guo, Yufeng Meyer, Jannik C. Wang, Haomin State Key Laboratory of Functional Materials for Informatics Shanghai Institute of Microsystem and Information Technology Chinese Academy of Sciences Shanghai200050 China Center of Materials Science and Optoelectronics Engineering University of Chinese Academy of Sciences Beijing100049 China School of Materials Science and Physics China University of Mining and Technology Xuzhou221116 China Faculty of Physics University of Vienna Vienna1090 Austria School of Chemistry Trinity College Dublin Dublin 2 Ireland State Key Laboratory of Mechanics and Control of Mechanical Structures MOE Key Laboratory for Intelligent Nano Materials and Devices College of Aerospace Engineering Nanjing University of Aeronautics and Astronautics Nanjing210016 China School of Physical Science and Technology ShanghaiTech University Shanghai201210 China Key Laboratory of Optoelectronic Technology Systems Ministry of Education International R & D Center of Micro-Nano Systems New Materials Technology Chongqing University Chongqing400044 China University of Antwerp Antwerp2020 Belgium National Institute for Materials Science Tsukuba305-0044 Japan Institute for Applied Physics and Natural Medical Sciences Institute University of Tübingen Tübingen72076 Germany

Hexagonal boron nitride (h-BN) has long been recognized as an ideal substrate for electronic devices due to its dangling-bond-free surface, insulating nature and thermal/chemical stability. These properties of the h-BN multilayer are mainly determined by its lattice structure. Therefore, to analyse the lattice structure and orientation of h-BN crystals becomes important. Here, the stacking order and wrinkles of h-BN are investigated by transmission electron microscopy (TEM). It is experimentally confirmed that the layers in the h-BN flakes are arranged in the AA′ stacking. The wrinkles in a form of threefold network throughout the h-BN crystal are oriented along the armchair direction, and their formation mechanism was further explored by molecular dynamics simulations. Our findings provide a deep insight about the microstructure of h-BN and shed light on the structural design/electronic modulations of two-dimensional crystals. © 2021, CC BY-NC-ND.

关键词： III-V semiconductors

来源：评论

学校读者我要写书评

暂无评论

Research on Answer Extraction for Automatic Question Answering System

Research on Answer Extraction for Automatic Question Answeri...

引用

作者： Tiantian Wu Hongzhi Yu Fucheng Wan Fangtao Yang Key Laboratory of China's Ethnic Languages and Information Technology of Ministry of Education Northwest Minzu University Key Laboratory of China's Ethnic Languages and Intelligent Processing of Gansu Province Northwest Minzu University

The question answering system as a hot issue in the field of NLP study,and answer extraction as the core part of the automatic question answering system,its effect is good or bad is directly related to the performance of question answering system,this paper introduces the technical process of answer extraction module and the use of pattern matching technology,gradually shift from the traditional pattern matching technology to the machine learning method of pattern matching technology,and also calculate the questions and answers in the answer extraction module are several methods of *** the performance of the automatic question answering system depends on the algorithm used for answer extraction to a large extent,the selection of the algorithm for answer extraction is also crucial,and the accuracy of the answer returned to the user can be improved by improving the algorithm.

关键词： Automatic Question Answering Answer Extraction Pattern Matching Similarity

来源：评论

学校读者我要写书评

暂无评论

Feature subset discernibility evaluation method for upper limb rehabilitation training based on the discernibility of relative distance 15

Feature subset discernibility evaluation method for upper li...

引用

15th IEEE International Conference on Mechatronics and Automation, ICMA 2018

作者： Zuo, Guoyu Xu, Zhaokun Lu, Jiahao Faculty of Information Technology Beijing University of Technology Beijing1000124 China Beijing Key Laboratory of Computing Intelligence and Intelligent Systems Beijing1000124 China

ISBN: (纸本)9781538660720

In this paper, a feature subset discernibility hybrid evaluation method based on the discernibility of relative distance and support vector machine (DRD-SVM) is proposed for the feature selection problem of the upper limb rehabilitation training motion of Brunnstrom 4-5 stage patients, in which a relative distance is introduced into evaluating the discernibility between classes considering the joint effect of both candidate and selected features. First, a feature subset search strategy is used to search a set of candidate feature subsets. Then the DRD is used to evaluate the candidate feature subsets, the best subset is selected as a new selected feature subset, and the feature subset with the best performance of SVM classification is selected as the optimal feature subset. Finally, feature selection experiment was carried out on upper limb routine rehabilitation training samples of the Brunnstrom 4-5 stage. The experimental results shows that, compared with the F-score method and the DFS one, the proposed method can obtain the feature subsets with higher accuracy and smaller feature dimension, which improves its effectiveness and feasibility. © 2018 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Diesel and bi-fuel probe vehicles noise emission assessment as a function of speed, gear and RPM

引用

AIP Conference Proceedings 2024年第1期3084卷

作者： Antonio Pascale Aurora Mascolo Domenico Rossi Simona Mancini Margarida C. Coelho Claudio Guarnaccia Department of Mechanical Engineering / Centre for Mechanical Technology and Automation (TEMA) University of Aveiro Campus Universitário de Santiago 3810-193 Aveiro – Portugal & LASI - Intelligent Systems Associate Laboratory Portugal Department of Civil Engineering Applied Physics Engineering Lab University of Salerno Via Giovanni Paolo II 132 Fisciano I-84084 Italy Department of Information and Electric Engineering and Applied Mathematics University of Salerno via Giovanni Paolo II 132 84084 Fisciano SA Italy

Nowadays, road traffic noise pollution still represents a considerable problem in Europe despite the efforts gathered by the scientific community, local authorities, and policymakers to tackle this issue. Research mostly focused on the development of Road Traffic Noise Models (RTNMs) and Noise Emission Models (NEMs) to forecast respectively traffic noise levels in terms of equivalent continuous sound levels and single-vehicle noise emissions in terms of sound power level. However, although the validity of such models was demonstrated in the literature, most of them are not able to deeply characterize the source in terms of motorization, engaged gear, and in idling conditions (i.e., a vehicle stopped in neutral gear with the engine running). Therefore, in this paper, an approach (based on standards and regulations) to obtain noise emission curves as a function of speed, motorization, and engaged gear, and a mathematical relationship between the sound pressure level (measured at the exhaust tube pipe) and the engine Revolution Per Minute (RPM) will be presented. Both methods are based on regressions (non-linear and linear, respectively). Two probe vehicles with different motorization (one diesel and one bi-fuel) were adopted to this end. Results showed that there exists a variability among the noise emission curves based on motorization information and engaged gear (at low speeds for manual transmission vehicles). Moreover, exhaust noise emissions, which are a function of motorization, linearly depend on RPM. All the obtained regressions are characterized by high determination coefficients (above 0.9) confirming the goodness of the presented approaches.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Weakly Supervised Object Detection With Segmentation Collaboration

Weakly Supervised Object Detection With Segmentation Collabo...

引用

International Conference on Computer Vision (ICCV)

作者： Xiaoyan Li Meina Kan Shiguang Shan Xilin Chen Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing Technology CAS Beijing China University of Chinese Academy of Sciences Beijing China Peng Cheng Laboratory Shenzhen China

ISBN: (数字)9781728148038

ISBN: (纸本)9781728148045

Weakly supervised object detection aims at learning precise object detectors, given image category labels. In recent prevailing works, this problem is generally formulated as a multiple instance learning module guided by an image classification loss. The object bounding box is assumed to be the one contributing most to the classification among all proposals. However, the region contributing most is also likely to be a crucial part or the supporting context of an object. To obtain a more accurate detector, in this work we propose a novel end-to-end weakly supervised detection approach, where a newly introduced generative adversarial segmentation module interacts with the conventional detection module in a collaborative loop. The collaboration mechanism takes full advantages of the complementary interpretations of the weakly supervised localization task, namely detection and segmentation tasks, forming a more comprehensive solution. Consequently, our method obtains more precise object bounding boxes, rather than parts or irrelevant surroundings. Expectedly, the proposed method achieves an accuracy of 53.7% on the PASCAL VOC 2007 dataset, outperforming the state-of-the-arts and demonstrating its superiority for weakly supervised object detection.

关键词： Proposals Object detection Feature extraction Collaboration Task analysis Detectors Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Experimental Quantification of Coherence of a Tunable Quantum Detector

引用

Physical Review Letters 2020年第6期125卷 060404-060404页

作者： Huichao Xu Feixiang Xu Thomas Theurer Dario Egloff Zi-Wen Liu Nengkun Yu Martin B. Plenio Lijian Zhang National Laboratory of Solid State Microstructures Key Laboratory of Intelligent Optical Sensing and Manipulation College of Engineering and Applied Sciences and Collaborative Innovation Center of Advanced Microstructures Nanjing University Nanjing 210093 China Institute of Theoretical Physics and IQST Universität Ulm Albert-Einstein-Allee 11 D-89069 Ulm Germany Technical University Dresden Institute of Theoretical Physics D-01062 Dresden Germany Perimeter Institute for Theoretical Physics Waterloo Ontario N2L 2Y5 Canada Center for Theoretical Physics Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA Centre for Quantum Computation and Intelligent Systems Faculty of Engineering and Information Technology University of Technology Sydney Sydney New South Wales 2007 Australia

Quantum coherence is a fundamental resource that quantum technologies exploit to achieve performance beyond that of classical devices. A necessary prerequisite to achieve this advantage is the ability of measurement devices to detect coherence from the measurement statistics. Based on a recently developed resource theory of quantum operations, here we quantify experimentally the ability of a typical quantum-optical detector, the weak-field homodyne detector, to detect coherence. We derive an improved algorithm for quantum detector tomography and apply it to reconstruct the positive-operator-valued measures of the detector in different configurations. The reconstructed positive-operator-valued measures are then employed to evaluate how well the detector can detect coherence using two computable measures. As the first experimental investigation of quantum measurements from a resource theoretical perspective, our work sheds new light on the rigorous evaluation of the performance of a quantum measurement apparatus.

关键词： Optical quantum information processing Quantum coherence & coherence measures Quantum tomography Resource theories

来源：评论

学校读者我要写书评

暂无评论

Application of Railway Passenger Flow Statistics Based on Mask R-CNN

引用

IOP Conference Series: Materials Science and Engineering 2020年第7期768卷

作者： Di Meng Shangzhi Xu Information Processing and Intelligent Transportation System Laboratory Department of Information and Communication Engineering Tongji University Shanghai 201804 China The Center of National Railway Intelligent Transportation System Engineering and Technology China

This paper discusses and implements the application of Mask R-CNN in railway passenger flow statistics. Mask R-CNN is a target detector based on candidate region, combining deep learning, neural network and image recognition technology, and has a wide range of application value in the field of intelligent *** paper focuses on the algorithm named Mask-Flow that Mask R-CNN uses the same color for tracking on the same target, and counts the number of pedestrians. And the passenger flow statistics are compared with Retina-Net in the railway scene of the dense crowd, which proves the accuracy of the method in the paper.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient Large-Scale Multi-graph Similarity Search Using MapReduce 16th

Efficient Large-Scale Multi-graph Similarity Search Using Ma...

引用

16th Web information systems and Applications Conference, WISA 2019

作者： Pang, Jun Yu, Minghe Gu, Yu School of Computer Science and Technology Wuhan University of Science and Technology WuhanHubei430065 China Guangdong Province Key Laboratory of Popular High Performance Computers Shenzhen University Shenzhen518061 China Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System WuhanHubei430065 China School of Computer Science and Engineering Northeastern University ShenyangLiaoning110819 China College of Software Northeastern University ShenyangLiaoning110819 China

ISBN: (纸本)9783030309510

A multi-graph is a set consisting of multiple graphs. Multi-graph similarity search aims to find the multi-graphs similar to the query multi-graphs from the multi-graph datasets. It plays important role in a wide range of application fields, such as finding similar drugs, searching similar molecule groups and so on. However, existing algorithms of multi-graph similarity search are memory-based algorithms, which are not suitable for the large amount of multi-graph scenarios. In this paper, we propose a parallel algorithm based on the MapReduce programming model to solve the problem of the large-scale multi-graph similarity search. Our proposed algorithm consists of two MapReduce jobs, one for indexing and the other for filtering and validation. Specially, we adapt the localization strategy to further improve the performance of our algorithm, which not only reduces the communication cost, but also mitigates the load imbalance. Extensive experimental results show that our algorithm is effective and efficient. © 2019, Springer Nature Switzerland AG.

关键词： MapReduce

来源：评论

学校读者我要写书评

暂无评论

An operation network for abstractive sentence compression 27

An operation network for abstractive sentence compression

引用

27th International Conference on Computational Linguistics, COLING 2018

作者： Yu, Naitong Zhang, Jie Huang, Minlie Zhu, Xiaoyan State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and Technology Dept. of Computer Science and Technology Tsinghua University Beijing China Microsoft Search Technology Center

ISBN: (纸本)9781948087506

Sentence compression condenses a sentence while preserving its most important contents. Delete-based models have the strong ability to delete undesired words, while generate-based models are able to reorder or rephrase the words, which are more coherent to human sentence compression. In this paper, we propose Operation Network, a neural network approach for abstractive sentence compression, which combines the advantages of both delete-based and generate-based sentence compression models. The central idea of Operation Network is to model the sentence compression process as an editing procedure. First, unnecessary words are deleted from the source sentence, then new words are either generated from a large vocabulary or copied directly from the source sentence. A compressed sentence can be obtained by a series of such edit operations (delete, copy and generate). Experiments show that Operation Network outperforms state-of-the-art baselines. © 2018 COLING 2018 - 27th International Conference on Computational Linguistics, Proceedings. All rights reserved.

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：