检索结果-内蒙古大学图书馆

Research Square 2021年

作者： Cuesta-Lazaro, Carolina Quera-Bofarull, Arnau Aylett-Bullock, Joseph Lawrence, Bryan N. Fong, Kevin Icaza-Lizaola, Miguel Sedgewick, Aidan Truong, Henry Vernon, Ian Williams, Julian Pagel, Christina Krauss, Frank Institute for Data Science Durham University Durham United Kingdom Institute for Computational Cosmology Durham University Durham United Kingdom Institute for Particle Physics Phenomenology Durham University Durham United Kingdom National Centre for Atmospheric Science University of Reading Reading United Kingdom Departments of Meteorology and Computer Science University of Reading Reading United Kingdom Department of Science Technology Engineering and Public Policy University College London London United Kingdom Department of Anaesthesia University College London Hospital London United Kingdom Centre for Extragalactic Astronomy Durham University Durham United Kingdom Department of Mathematical Sciences Durham University DurhamDH1 3LE United Kingdom Institute for Hazard Risk & Resilience Durham University DurhamDH1 3LE United Kingdom Clinical Operational Research Unit University College London United Kingdom

With high levels of the Delta variant of COVID-19 circulating in England during September 2021, schools are set to reopen with few school-based non-pharmaceutical interventions (NPIs). In this paper, we present simulation results obtained from the individual-based model, JUNE, for English school opening after a prior vaccination campaign using an optimistic set of assumptions about vaccine efficacy and the likelihood of prior-reinfection. We take a scenario-based approach to modelling potential interventions to assess relative changes rather than real-world forecasts. Specifically, we assess the effects of vaccinating those aged 16-17, those aged 12-17, and not vaccinating children at all relative to only vaccinating the adult population, addressing what might have happened had the UK began teenage vaccinations earlier. Vaccinating children in the 12-15 age group would have had a significant impact on the course of the epidemic, saving thousands of lives overall in these simulations. In the absence of such a vaccination campaign our simulations show there could still be a significant positive impact on the epidemic (fewer cases, fewer deaths) by continuing NPI strategies in schools. Our analysis suggests that the best results in terms of lives saved are likely derived from a combination of the now planned vaccination campaign and NPIs in schools. © 2021, CC BY.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Minimalistic attacks: How little it takes to fool deep reinforcement learning policies

arXiv

引用

arXiv 2019年

作者： Qu, Xinghua Sun, Zhu Ong, Yew Soon Wei, Pengfei Gupta, Abhishek Computational Intelligence Lab School of Computer Science and Engineering Nanyang Technological University Singapore639798 Singapore School of Electrical and Electronic Engineering Nanyang Technological University Singapore639798 Singapore Data Science and Artificial Intelligence Research Centre School of Computer Science and Engineering Nanyang Technological University Singapore639798 Singapore Department of Computer Science National University of Singapore Singapore119077 Singapore A*STAR Singapore138634 Singapore

—Recent studies have revealed that neural network based policies can be easily fooled by adversarial examples. However, while most prior works analyze the effects of perturbing every pixel of every frame assuming white-box policy access, in this paper we take a more restrictive view towards adversary generation - with the goal of unveiling the limits of a model’s vulnerability. In particular, we explore minimalistic attacks by defining three key settings: (1) black-box policy access: where the attacker only has access to the input (state) and output (action probability) of an RL policy;(2) fractional-state adversary: where only several pixels are perturbed, with the extreme case being a single-pixel adversary;and (3) tactically-chanced attack: where only significant frames are tactically chosen to be attacked. We formulate the adversarial attack by accommodating the three key settings, and explore their potency on six Atari games by examining four fully trained state-of-the-art policies. In Breakout, for example, we surprisingly find that: (i) all policies showcase significant performance degradation by merely modifying 0.01% of the input state, and (ii) the policy trained by DQN is totally deceived by perturbation to only 1% frames. Copyright © 2019, The Authors. All rights reserved.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Decoding of visual-related information from the human EEG using an end-to-end deep learning approach

arXiv

引用

arXiv 2019年

作者： Yang, Lingling Chan, Leanne Lai Hang Lu, Yao University of Minnesota Department of Electrical Engineering City University of Hong Kong HKSAR China School of Data and Computer Science and Guangdong Province Key Laboratory of Computational Science Sun Yat-Sen University Guangzhou China

There is increasing interest in using deep learning approach for EEG analysis as there are still rooms for the improvement of EEG analysis in its accuracy. Convolutional long short-term (CNNLSTM) has been successfully applied in time series data with spatial structure through end-to-end learning. Here, we proposed a CNNLSTM based neural network architecture termed EEG-CNNLSTMNet for the classification of EEG signals in response to grating stimuli with different spatial frequencies. EEG-CNNLSTMNet comprises two convolutional layers and one bidirectional long short-term memory (LSTM) layer. The convolutional layers capture local temporal characteristics of the EEG signal at each channel as well as global spatial characteristics across channels, while the LSTM layer extracts long-term temporal dependency of EEG signals. Our experiment showed that EEG-CNNLSTMNet performed much better at EEG classification than a traditional machine learning approach, i.e. a support vector machine (SVM) with features. Additionally, EEG-CNNLSTMNet outperformed EEGNet, a state-of-art neural network architecture for the intra-subject case. We infer that the underperformance when using an LSTM layer in the inter-subject case is due to long-term dependency characteristics in the EEG signal that vary greatly across subjects. Moreover, the inter-subject fine-tuned classification model using very little data of the new subject achieved much higher accuracy than that trained only on the data from the other subjects. Our study suggests that the fine-tuned inter-subject model can be a potential end-to-end EEG analysis method considering both the accuracy and the required training data of the new subject. Copyright © 2019, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Serial-EMD: Fast empirical mode decomposition method for multi-dimensional signals based on serialization

arXiv

引用

arXiv 2021年

作者： Zhang, Jin Feng, Fan Marti-Puig, Pere Caiafa, Cesar F. Sun, Zhe Duan, Feng Solé-Casals, Jordi College of Computer Science Nankai University Tianjin300071 China College of Artificial Intelligence Nankai University Tianjin300350 China Data and Signal Processing Group University of Vic—Central University of Catalonia Catalonia Vic08500 Spain Instituto Argentino de Radioastronomía CONICET CCT La Plata CIC-PBA UNLP V. Elisa1894 Argentina Tensor Learning Team Center for Advanced Intelligence Project RIKEN Tokyo103-0027 Japan Computational Engineering Applications Unit Head Office for Information Systems and Cybersecurity RIKEN Saitama351-0198 Japan

Empirical mode decomposition (EMD) has developed into a prominent tool for adaptive, scale-based signal analysis in various fields like robotics, security and biomedical engineering. Since the dramatic increase in amount of data puts forward higher requirements for the capability of real-time signal analysis, it is difficult for existing EMD and its variants to trade off the growth of data dimension and the speed of signal analysis. In order to decompose multi-dimensional signals at a faster speed, we present a novel signal-serialization method (serial-EMD), which concatenates multi-variate or multi-dimensional signals into a one-dimensional signal and uses various one-dimensional EMD algorithms to decompose it. To verify the effects of the proposed method, synthetic multi-variate time series, artificial 2D images with various textures and real-world facial images are tested. Compared with existing multi-EMD algorithms, the decomposition time becomes significantly reduced. In addition, the results of facial recognition with Intrinsic Mode Functions (IMFs) extracted using our method can achieve a higher accuracy than those obtained by existing multi-EMD algorithms, which demonstrates the superior performance of our method in terms of the quality of IMFs. Furthermore, this method can provide a new perspective to optimize the existing EMD algorithms, that is, transforming the structure of the input signal rather than being constrained by developing envelope computation techniques or signal decomposition methods. In summary, the study suggests that the serial-EMD technique is a highly competitive and fast alternative for multi-dimensional signal analysis. Copyright © 2021, The Authors. All rights reserved.

关键词： Signal analysis

来源：评论

学校读者我要写书评

暂无评论

An Open-Source Knowledge Graph Ecosystem for the Life sciences

arXiv

引用

arXiv 2023年

作者： Callahan, Tiffany J. Tripodi, Ignacio J. Stefanski, Adrianne L. Cappelletti, Luca Taneja, Sanya B. Wyrwa, Jordan M. Casiraghi, Elena Matentzoglu, Nicolas A. Reese, Justin Silverstein, Jonathan C. Hoyt, Charles Tapley Boyce, Richard D. Malec, Scott A. Unni, Deepak R. Joachimiak, Marcin P. Robinson, Peter N. Mungall, Christopher J. Cavalleri, Emanuele Fontana, Tommaso Valentini, Giorgio Mesiti, Marco Gillenwater, Lucas A. Santangelo, Brook Vasilevsky, Nicole A. Hoehndorf, Robert Bennett, Tellen D. Ryan, Patrick B. Hripcsak, George Kahn, Michael G. Bada, Michael Baumgartner, William A. Hunter, Lawrence E. Computational Bioscience Program University of Colorado Anschutz Medical Campus AuroraCO80045 United States Department of Biomedical Informatics Columbia University Irving Medical Center New YorkNY10032 United States Computer Science Department Interdisciplinary Quantitative Biology University of Colorado Boulder BoulderCO80301 United States AnacletoLab Computer Science Department University of Milan 20122 Italy Intelligent Systems Program University of Pittsburgh PittsburghPA15260 United States Department of Physical Medicine and Rehabilitation School of Medicine University of Colorado Anschutz Medical Campus AuroraCO80045 United States Division of Environmental Genomics and Systems Biology Lawrence Berkeley National Laboratory BerkeleyCA94720 United States Semanticly Ltd Athens Greece Department of Biomedical Informatics University of Pittsburgh School of Medicine PittsburghPA15206 United States Laboratory of Systems Pharmacology Harvard Medical School BostonMA02115 United States Division of Translational Informatics University of New Mexico School of Medicine AlbuquerqueNM87131 United States SIB Swiss Institute of Bioinformatics Basel Switzerland Berlin Institute of Health at Charité-Universitatsmedizin Berlin10117 Germany ELLIS European Laboratory for Learning and Intelligent Systems Germany Department of Biomedical Informatics University of Colorado School of Medicine AuroraCO80045 United States Data Collaboration Center Critical Path Institute 1840 E River Rd. Suite 100 TucsonAZ85718 United States Computer Electrical and Mathematical Sciences & Engineering Division Computational Bioscience Research Center King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia Department of Pediatrics University of Colorado School of Medicine AuroraCO80045 United States Janssen Research and Development RaritanNJ08869 United States Division of General Internal Medicine University of Colorado School of Medicine A

Translational research requires data at multiple scales of biological organization. Advancements in sequencing and multi-omics technologies have increased the availability of these data, but researchers face significant integration challenges. Knowledge graphs (KGs) are used to model complex phenomena, and methods exist to construct them automatically. However, tackling complex biomedical integration problems requires flexibility in the way knowledge is modeled. Moreover, existing KG construction methods provide robust tooling at the cost of fixed or limited choices among knowledge representation models. PheKnowLator (Phenotype Knowledge Translator) is a semantic ecosystem for automating the FAIR (Findable, Accessible, Interoperable, and Reusable) construction of ontologically grounded KGs with fully customizable knowledge representation. The ecosystem includes KG construction resources (e.g., data preparation APIs), analysis tools (e.g., SPARQL endpoints and abstraction algorithms), and benchmarks (e.g., prebuilt KGs and embeddings). We evaluated the ecosystem by systematically comparing it to existing open-source KG construction methods and by analyzing its computational performance when used to construct 12 large-scale KGs. With flexible knowledge representation, PheKnowLator enables fully customizable KGs without compromising performance or usability. © 2023, CC BY.

关键词： Ecosystems

来源：评论

学校读者我要写书评

暂无评论

GIscience in the Era of Artificial Intelligence: A Research Agenda Towards Autonomous GIS

arXiv

引用

arXiv 2025年

作者： Li, Zhenlong Ning, Huan Gao, Song Janowicz, Krzysztof Li, Wenwen Arundel, Samantha T. Yang, Chaowei Bhaduri, Budhendra Wang, Shaowen Zhu, A. Xing Gahegan, Mark Shekhar, Shashi Ye, Xinyue McKenzie, Grant Cervone, Guido Hodgson, Michael E. Geoinformation and Big Data Research Lab Department of Geography The Pennsylvania State University University ParkPA United States Department of Geography University of Wisconsin – Madison WI United States STKO Lab Department of Geography and Regional Research University of Vienna Vienna Austria Spatial Analysis Research Center School of Geographical Sciences and Urban Planning Arizona State University AZ United States Center of Excellence for Geospatial Information Science U.S. Geological Survey VA United States NSF Spatiotemporal Innovation Center Department of Geography & Geoinformation Science George Mason University VA United States TN United States CyberGIS Center for Advanced Digital and Spatial Studies Department of Geography and Geographic Information Science University of Illinois Urbana-Champaign IL United States School of Computer Science University of Auckland New Zealand Department of Computer Science & Engineering University of Minnesota MN United States Department of Landscape Architecture & Urban Planning Center for Geospatial Sciences Applications & Technology Texas A&M University TX United States Platial Analysis Lab Department of Geography McGill University Quebec Canada Institute for Computational and Data Sciences Department of Geography The Pennsylvania State University University ParkPA United States Department of Geography University of South Carolina SC United States

The advent of generative AI exemplified by large language models (LLMs) opens new ways to represent and compute geographic information and transcends the process of geographic knowledge production, driving geographic information systems (GIS) towards autonomous GIS. Leveraging LLMs as the decision core, autonomous GIS can independently generate and execute geoprocessing workflows to perform spatial analysis. In this vision paper, we further elaborate on the concept of autonomous GIS and present a conceptual framework that defines its five autonomous goals, five autonomous levels, five core functions, and three operational scales. We demonstrate how autonomous GIS could perform geospatial data retrieval, spatial analysis, and map making with four proof-of-concept GIS agents. We conclude by identifying critical challenges and future research directions, including fine-tuning and self-growing decision-cores, autonomous modeling, and examining the societal and practical implications of autonomous GIS. By establishing the groundwork for a paradigm shift in GIscience, this paper envisions a future where GIS moves beyond traditional workflows to autonomously reason, derive, innovate, and advance geospatial solutions to pressing global challenges. Meanwhile, as we design and deploy increasingly intelligent geospatial systems, we carry a responsibility to ensure they are developed in socially responsible ways, serve the public good, and support the continued value of human geographic insight in an AI-augmented future. © 2025, CC BY.

关键词： Image retrieval

来源：评论

学校读者我要写书评

暂无评论

Handling of uncertainty in medical data using machine learning and probability theory techniques: A review of 30 years (1991-2020)

arXiv

引用

arXiv 2020年

作者： Alizadehsani, Roohallah Roshanzamir, Mohamad Hussain, Sadiq Khosravi, Abbas Koohestani, Afsaneh Zangooei, Mohammad Hossein Abdar, Moloud Beykikhoshk, Adham Shoeibi, Afshin Zare, Assef Panahiazar, Maryam Nahavandi, Saeid Srinivasan, Dipti Atiya, Amir F. Acharya, U. Rajendra Deakin University Geelong Australia Department of Engineering Fasa Branch Islamic Azad University Post Box No 364 Fasa Fars*** Iran System Administrator Dibrugarh University Assam786004 India University of Texas Dallas United States Applied Artificial Intelligence Institute Deakin University Geelong Australia Computer Engineering Department Ferdowsi University of Mashhad Mashhad Iran Faculty of Electrical and Computer Engineering Biomedical Data Acquisition Lab K. N. Toosi University of Technology Tehran Iran Faculty of Electrical Engineering Gonabad Branch Islamic Azad University Gonabad Iran Institute for Computational Health Sciences University of California San Francisco United States Dept. of Electrical and Computer Engineering National University of Singapore Singapore117576 Singapore Department of Computer Engineering Faculty of Engineering Cairo University Cairo12613 Egypt Department of Electronics and Computer Engineering Ngee Ann Polytechnic Singapore Singapore Department of Biomedical Engineering School of Science and Technology Singapore University of Social Sciences Singapore Department of Bioinformatics and Medical Engineering Asia University Taiwan

Understanding data and reaching valid conclusions are of paramount importance in the present era of big data. Machine learning and probability theory methods have widespread application for this purpose in different fields. One critically important yet less explored aspect is how data and model uncertainties are captured and analyzed. Proper quantification of uncertainty provides valuable information for optimal decision making. This paper reviewed related studies conducted in the last 30 years (from 1991 to 2020) in handling uncertainties in medical data using probability theory and machine learning techniques. Medical data is more prone to uncertainty due to the presence of noise in the data. So, it is very important to have clean medical data without any noise to get accurate diagnosis. The sources of noise in the medical data need to be known to address this issue. Based on the medical data obtained by the physician, diagnosis of disease, and treatment plan are prescribed. Hence, the uncertainty is growing in healthcare and there is limited knowledge to address these problems. We have little knowledge about the optimal treatment methods as there are many sources of uncertainty in medical science. Our findings indicate that there are few challenges to be addressed in handling the uncertainty in medical raw data and new models. In this work, we have summarized various methods employed to overcome this problem. Nowadays, application of novel deep learning techniques to deal such uncertainties have significantly increased. Copyright © 2020, The Authors. All rights reserved.

关键词： Monte Carlo methods

来源：评论

学校读者我要写书评

暂无评论

OpenMM 8: Molecular Dynamics Simulation with Machine Learning Potentials

arXiv

引用

arXiv 2023年

作者： Eastman, Peter Galvelis, Raimondas Peláez, Raúl P. Abreu, Charlles R.A. Farr, Stephen E. Gallicchio, Emilio Gorenko, Anton Henry, Michael M. Hu, Frank Huang, Jing Krämer, Andreas Michel, Julien Mitchell, Joshua A. Pande, Vijay S. Rodrigues, João P.G.L.M. Rodriguez-Guerra, Jaime Simmonett, Andrew C. Singh, Sukrit Swails, Jason Turner, Philip Wang, Yuanqing Zhang, Ivy Chodera, John D. De Fabritiis, Gianni Markland, Thomas E. Department of Chemistry Stanford University StanfordCA94305 United States Acellera Labs C Dr Trueta 183 Barcelona08005 Spain C Dr. Aiguader 88 Barcelona08003 Spain Chemical Engineering Department School of Chemistry Federal University of Rio de Janeiro Rio de Janeiro68542 Brazil Redesign Science Inc. 180 Varick St. New YorkNY10014 United States EaStCHEM School of Chemistry University of Edinburgh EH9 3FJ United Kingdom Department of Chemistry and Biochemistry Brooklyn College The City University of New York NY United States Ph.D. Program in Chemistry Ph.D. Program in Biochemistry The Graduate Center of the City University of New York New YorkNY United States Stream HPC Koningin Wilhelminaplein 1 - 40601 Amsterdam1062 HG Netherlands Computational and Systems Biology Program Sloan Kettering Institute Memorial Sloan Kettering Cancer Center New YorkNY10065 United States Key Laboratory of Structural Biology of Zhejiang Province School of Life Sciences Westlake University 18 Shilongshan Road Zhejiang Hangzhou310024 China Department of Mathematics and Computer Science Freie Universität Berlin Arnimallee 12 Berlin14195 Germany The Open Force Field Initiative Open Molecular Software Foundation DavisCA95616 United States Andreessen Horowitz 2865 Sand Hill Rd Menlo ParkCA94025 United States Department of Structural Biology Stanford University StanfordCA94305 United States Charité Universitätsmedizin Berlin In silico Toxicology and Structural Bioinformatics Virchowweg 6 Berlin10117 Germany Laboratory of Computational Biology National Heart Lung and Blood Institute National Institutes of Health BethesdaMD20892 United States Entos Inc. 9310 Athena Circle La Jolla CA92037 United States College of Engineering Virginia Polytechnic Institute State University BlacksburgVA24061 United States Simons Center for Computational Physical Chemistry Center for Data Science New York University 24 Waverly Place New YorkNY10004 United States T

Machine learning plays an important and growing role in molecular simulation. The newest version of the OpenMM molecular dynamics toolkit introduces new features to support the use of machine learning potentials. Arbitrary PyTorch models can be added to a simulation and used to compute forces and energy. A higher-level interface allows users to easily model their molecules of interest with general purpose, pretrained potential functions. A collection of optimized CUDA kernels and custom PyTorch operations greatly improves the speed of simulations. We demonstrate these features on simulations of cyclin-dependent kinase 8 (CDK8) and the green fluorescent protein (GFP) chromophore in water. Taken together, these features make it practical to use machine learning to improve the accuracy of simulations at only a modest increase in cost. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Vec2SPARQL: Integrating SPARQL queries and knowledge graph embeddings 11

Vec2SPARQL: Integrating SPARQL queries and knowledge graph e...

引用

11th International Conference Semantic Web Applications and Tools for Life sciences, SWAT4LS 2018

作者： Kulmanov, Maxat Kafkas, Senay Karwath, Andreas Malic, Alexander Gkoutos, Georgios V Dumontier, Michel Hoehndorf, Robert Computer Electrical and Mathematical Science and Engineering Division Computational Bioscience Research Center King Abdullah University of Science and Technology Thuwal23955 Saudi Arabia Centre for Computational Biology University of Birmingham Birmingham United Kingdom Institute of Data Science Maastricht University Maastricht Netherlands

Recent developments in machine learning have led to a rise of large number of methods for extracting features from structured data. The features are represented as vectors and may encode for some semantic aspects of data. They can be used in a machine learning models for different tasks or to compute similarities between the entities of the data. SPARQL is a query language for structured data originally developed for querying Resource Description Framework (RDF) data. It has been in use for over a decade as a standardized NoSQL query language. Many different tools have been developed to enable data sharing with SPARQL. For example, SPARQL endpoints make your data interoperable and available to the world. SPARQL queries can be executed across multiple endpoints. We have developed a Vec2SPARQL, which is a general framework for integrating structured data and their vector space representations. Vec2SPARQL allows jointly querying vector functions such as computing similarities (cosine, correlations) or classifications with machine learning models within a single SPARQL query. We demonstrate applications of our approach for biomedical and clinical use cases. Our source code is freely available at https://***/bio-ontology-research-group/vec2sparql and we make a Vec2SPARQL endpoint available at http://***/. © 2018 CEUR Workshop Proceedings. All rights reserved.

关键词： Vector spaces

来源：评论

学校读者我要写书评

暂无评论

Accommodating Information Priority Model in Cloudlet Environment 6th

引用

6th Australasian Symposium on Service Research and Innovation, ASSRI 2017

作者： Geumpana, Teuku Aulia Rabhi, Fethi Zhu, Liming School of Computer Science and Engineering The University of New South Wales Sydney2052 Australia Data61 | CSIRO Software and Computational Systems Sydney2015 Australia

ISBN: (纸本)9783319765860

Massive amounts of data during disaster situations require timely collection and analysis for the emergency team to mitigate the impact of the disaster under challenging social-technical conditions. The absence of Internet or its intermittent and bandwidth-constraint connection in disaster areas may exacerbate and disrupt the data collection process which may prevent some vital information to reach the control room in time for immediate response. Regardless the rare connection in the disaster area, there is a need to group information acquired during the response into a specific information model which accommodates different information priority levels. This is to establish a proper mechanism in transmitting higher prioritized information to the control room before other information. The purpose of this paper is to propose an information priority model and system architectures for data collection under challenging conditions in disaster areas. © 2018, Springer International Publishing AG.

关键词： data acquisition

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：