检索结果-内蒙古大学图书馆

您好，读者！请登录

内蒙古大学图书馆

首页
概况
党建
资源
服务
科研支持
- 论文收录引用证明
- 科技查新
知识产权
档案馆
帮助

咨询与建议

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

您的常用邮箱：*

您的手机号码：*

问题描述：

当前已输入0个字，您还可以输入200个字

全部搜索
期刊论文
图书
学位论文
标准
纸本馆藏
外文资源发现
数据库导航
超星发现

高级检索

时间限定

出版年份：

文献类型

图书期刊文献学位论文多媒体

馆藏选择

电子馆藏纸本馆藏

核心期刊

全部期刊 SCI 收录期刊 SSCI 收录期刊 EI 收录期刊 CSCD 收录期刊 CSSCI 收录期刊

语言

中文英文

文献类型

期刊文献图书学位论文标准纸本馆藏

帮助

文字说明：

T=题名（书名、题名），A=作者（责任者），K=主题词，P=出版物名称，PU=出版社名称，O=机构（作者单位、学位授予单位、专利申请人），L=中图分类号，C=学科分类号，U=全部字段，Y=年（出版发行年、学位年度、标准发布年）

检索规则说明：

AND代表“并且”；OR代表“或者”；NOT代表“不包含”；(注意必须大写,运算符两边需空一格)

检索范例：

范例一：(K=图书馆学 OR K=情报学) AND A=范并思 AND Y=1982-2016
范例二：P=计算机应用与软件 AND (U=C++ OR U=Basic) NOT K=Visual AND Y=2011-2016

分类表

所选分类

>> <<

限定检索结果

文献类型

6,061 篇 会议
909 篇 期刊文献
72 册 图书

馆藏范围

7,042 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

4,443 篇 工学
- 3,471 篇 计算机科学与技术...
- 2,907 篇 软件工程
- 1,321 篇 信息与通信工程
- 672 篇 控制科学与工程
- 478 篇 生物工程
- 436 篇 电气工程
- 434 篇 电子科学与技术（可...
- 329 篇 生物医学工程（可授...
- 305 篇 光学工程
- 271 篇 机械工程
- 233 篇 仪器科学与技术
- 211 篇 动力工程及工程热...
- 170 篇 化学工程与技术
- 160 篇 安全科学与工程
- 159 篇 交通运输工程
- 132 篇 建筑学
2,121 篇 理学
- 1,192 篇 数学
- 539 篇 物理学
- 496 篇 生物学
- 376 篇 统计学（可授理学、...
- 242 篇 系统科学
- 163 篇 化学
1,165 篇 管理学
- 697 篇 管理科学与工程(可...
- 517 篇 图书情报与档案管...
- 356 篇 工商管理
294 篇 医学
- 256 篇 临床医学
- 234 篇 基础医学(可授医学...
- 141 篇 公共卫生与预防医...
- 141 篇 药学(可授医学、理...
200 篇 法学
- 149 篇 社会学
145 篇 农学
117 篇 教育学
109 篇 经济学
25 篇 文学
16 篇 军事学
9 篇 艺术学
1 篇 哲学

主题

197 篇 accuracy
190 篇 feature extracti...
154 篇 computational mo...
145 篇 training
115 篇 computer science
114 篇 deep learning
111 篇 real-time system...
105 篇 data mining
101 篇 support vector m...
95 篇 servers
94 篇 predictive model...
93 篇 machine learning
87 篇 artificial intel...
84 篇 data models
83 篇 cloud computing
82 篇 internet of thin...
81 篇 protocols
79 篇 semantics
78 篇 computer archite...
78 篇 quality of servi...

机构

115 篇 dept. of compute...
85 篇 dept. of informa...
76 篇 institute of inf...
60 篇 dept. of informa...
56 篇 dept. of compute...
52 篇 dept. of compute...
44 篇 dept. of compute...
41 篇 dept. of compute...
37 篇 dept. of compute...
33 篇 dept. of compute...
28 篇 dept. of compute...
26 篇 dept. of compute...
22 篇 school of automa...
22 篇 graduate institu...
22 篇 dept. of compute...
20 篇 dept. of compute...
20 篇 dept. of compute...
19 篇 dept. of informa...
17 篇 dept. of compute...
17 篇 dept. of compute...

作者

46 篇 lorenzo bruzzone
25 篇 li kuan-ching
22 篇 jian-jhih kuo
20 篇 libertario demi
20 篇 yen-lin chen
19 篇 ren-hung hwang
18 篇 mei-hui wang
18 篇 chen yen-lin
18 篇 luca turchet
16 篇 francesca bovolo
16 篇 chang-shing lee
16 篇 hwang ren-hung
15 篇 kwon soonhak
15 篇 jia-ching wang
15 篇 hung yi-ping
14 篇 lee yue-shi
14 篇 bae hae-young
13 篇 horng shi-jinn
13 篇 watabe hirokazu
13 篇 wu eric hsiao-ku...

语言

6,733 篇 英文
224 篇 其他
60 篇 中文
19 篇 日文
6 篇 朝鲜文
1 篇 斯洛文尼亚文

检索条件"机构=Dept. of Computer Science& Information Engineering"

共 7042 条记录，以下是281-290 订阅

全选清除本页清除全部题录导出标记到"检索档案"

详细简洁

排序：

Generalised Active Learning With Annotation Quality Selection 33

Generalised Active Learning With Annotation Quality Selectio...

引用

33rd IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2023

作者： Lindqvist, Jakob Olmin, Amanda Svensson, Lennart Lindsten, Fredrik Chalmers University of Technology Dept. of Electrical Engineering Gothenburg Sweden Linköping University Dept. of Computer and Information Science Linköping Sweden

ISBN: (纸本)9798350324112

In this paper we promote a general formulation of active learning (AL), wherein the typically binary decision to annotate a point or not is extended to selecting the qualities with which the points should be annotated. By linking the annotation quality to the cost of acquiring the label, we can trade a lower quality for a larger set of training samples, which may improve learning for the same annotation cost. To investigate this AL formulation, we introduce a concrete criterion, based on the mutual information (MI) between model parameters and noisy labels, for selecting annotation qualities for the entire dataset, before any labels are acquired. We illustrate the usefulness of our formulation with examples for both classification and regression and find that MI is a good candidate for a criterion, but its complexity limits its usefulness. © 2023 IEEE.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

3D-CAN: A 3D Convolution Attention Network for Feature Extraction and Classification of Hyperspectral Images

3D-CAN: A 3D Convolution Attention Network for Feature Extra...

引用

2023 IEEE International Conference on Network, Multimedia and information Technology, NMITCON 2023

作者： Kanthi, Murali Rao, K. Venkateshwara Sekhar Reddy, L. Chandra Sarma, T. Hitendra Bhaskar, Nuthanakanti Vasundhara, Nam Telangana Hyderabad India Cmr College of Engineering and Technology Dept. of Computer Science and Engineering Telangana Hyderabad India Telangana Hyderabad India Vasavi College of Engineering Dept. of Information Technology Telangana Hyderabad India Cmr Technical Campus Dept. of Computer Science and Engineering Telangana Hyderabad India

ISBN: (纸本)9798350300826

In recent years, the deep learning method has gained widespread popularity in hyperspectral image classification (HSI) tasks because of remarkable feature extraction capability. However, conventional deep learning approaches often yield suboptimal classification performance due to the limited discrimination ability of the extracted features. Moreover, the challenge of obtaining high classification performance with a limited number of samples in hyperspectral images (HSIs) has emerged as a prominent research area of interest. To address this concern, this letter provides a novel 3D convolution attention network (3D-CAN) for feature extraction and HSIs classification. The attention mechanism employed in this context enables capturing both the long-range spatial dependencies of HSIs in the horizontal and vertical directions and the varying significance levels across distinct spectral channels. The experiments are executed on both the new data sets and the publicly accessible benchmark data sets to expose the efficiency and durability of the proposed model. Empirically, it has been demonstrated that the presented model succeeds over the remaining state-of-the-art approaches in terms of classification accuracy. © 2023 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Human Activity Recognition in Free Living Using a Wearable Sensor: A Two-Stage Approach 12

Human Activity Recognition in Free Living Using a Wearable S...

引用

12th International Conference on Awareness science and Technology, iCAST 2023

作者： Li, Tianhui Zhou, Xue Kokubun, Kota Chen, Wenxi Division of Information System The University of Aizu Aizu-Wakamatsu Japan Dept. of Computer and Information Systems The University of Aizu Aizu-Wakamatsu Japan Dept. of Computer Science and Engineering The University of Aizu Aizu-Wakamatsu Japan

ISBN: (纸本)9798350324693

A novel two-stage approach for human activity recognition (HAR) using a wearable sensor is proposed in this study. The goal is to achieve both coarse-level and fine-level recognition of activities. The study focuses on five common activities: desk-working, driving, walking, downstairs, and upstairs. Three-axis acceleration signals were collected using a wrist-worn accelerometer sensor, and the activities were annotated during the participant’s daily routines. The acceleration data were preprocessed through segmentation, feature extraction, and normalization. In the first s tage, a t raditional m achine learning model (decision tree) was used for coarse-level classification, distinguishing between activities with exercise and without exercise. In the second stage, deep learning models were employed for fine-level classification to recognize specific activities within the coarse classes. The performance of the models was evaluated using accuracy, sensitivity, and specificity. T he p roposed approach showed promising results with accuracies of 93.53%, 93.52%, 85.70%, 98.11%, and 85.91% respectively for desk-working, driving, walking, downstairs, and upstairs, demonstrating the effectiveness of the two-stage strategy for HAR. The proposed two-stage approach outperforms the one-stage method (decision tree classifier a nd d eep l earning m odel), p articularly i n imbalanced data, achieving a graceful equilibrium between sensitivity and specificity. T he s tudy h ighlights t he i mportance o f considering the complexity and variability of daily activities for accurate classification a nd r esource optimization. © 2023 IEEE.

关键词： Wearable sensors

来源：评论

学校读者我要写书评

暂无评论

A Multiple Branch Fusion Network for Feature Learning and Hyperspectral Image Classification 14

A Multiple Branch Fusion Network for Feature Learning and Hy...

引用

14th International Conference on Computing Communication and Networking Technologies, ICCCNT 2023

作者： Kanthi, Murali Rao, K. Venkateshwara Pavani K, Sudha Bhaskar, Nuthanakanti Bhukya, Shankar Nayak Hitendra Sarma, T. Cmr Technical Campus Telangana Hyderabad India CMR College of Engineering & Technology Dept. of Computer Science & Engineering Telangana Hyderabad India Dept. of Computer Science & Engineering Cmr Technical Campus Telangana Hyderabad India Vasavi College of Engineering Dept. of Information Technology Telangana Hyderabad India

ISBN: (纸本)9798350335095

Hyperspectral image (HSI) classification has made extensive use of deep learning approaches, specifically Convolutional Neural Network (CNN). CNN's ability to fully capture local contextual aspects of HSI makes it an excellent HSI classifier. Although, a significant factor in the poor classification performance is the scarcity of training data. Conventional CNN-based methods don't use the integrate-band relationships of HSI as much as they could because they typically employ 2D-CNNs to extract features. On the other hand, 3D-CNNs use a similar procedure to get both spectral and spatial data. This gets around the problem with 2D-CNNs, but it could mean that not enough features are extracted. As a solution, we offer a multi-scale 3D-CNN with multi branch feature fusion, which we call multi branch fusion CNN (MBF-CNN), for HSI classification. We begin by applying 3D-CNN on multiple scales to extract HSI features. After that, the multiple distinct characteristics are streamlined and combined together. The fused features then go through several fully connected layers to get the classification outcomes. The experiments are executed on both the new data sets and the publicly accessible benchmark data sets to expose the efficiency and durability of the proposed model. Empirically, it has been demonstrated that the presented model succeeds over the remaining state-of-the-art approaches in terms of classification accuracy. © 2023 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Movie Recommendation System Using Euclidean Distance

Movie Recommendation System Using Euclidean Distance

引用

2024 OPJU International Technology Conference on Smart Computing for Innovation and Advancement in Industry 4.0, OTCON 2024

作者： Khan, Muneeb Sami Hussain, Zunnurain Amaad, Muhammad Irtaza Hasan, Muhammad Zulkifl Khalid, Aqsa Chuhan, Saad Hussain Awan, Rimsha Yaqub, Muhammad Atif Ashraf, Farhan Information Technology University Department of Computer Engineering Lahore Pakistan Bahria University Lahore Campus Dept. of Computer Science Pakistan University of Central Punjab Faculty of Information Technology India National College of Business Administration and Economics Department of Computer Science Lahore Pakistan

ISBN: (纸本)9798350373783

In this study, we pivot from the commonly used cosine similarity to the Euclidean distance as the foundation for our film recommendation algorithm. Initially, data pertaining to users, films, and ratings are loaded using the panda's library. The algorithm then computes normalized ratings, tabulates the number of films in each genre, and visualizes this distribution. The frequency of ratings for each genre is similarly plotted. A dictionary is maintained to store movie IDs for each genre. These values are then integrated into the normalized rating matrix. For each genre, the normalized ratings of movies are cataloged in a lexicon. Using the normalized ratings matrix and movie IDs, a Data Frame is generated. Subsequently, the movie and ratings datasets are merged. A pivot table captures the number of films each user has watched per genre. The study also be happy with their purchase if these recommendations are tailored to their needs. As a result, the client would utilize this application once more. Because of the significant money gained determines the average number of films a user watches in each genre. This pivot table is refined to display movies in each genre that a user has watched, but only those exceeding the genre's average viewership. Instead of cosine similarity, the Euclidean distance is computed to gauge the similarity between movies, offering a fresh perspective on recommendation systems. © 2024 IEEE.

关键词： ContentBased Filtering Data Normalization Euclidean Distance Genre Analysis Machine Learning Algorithms Movie Recommendation System Personalized Recommendations Similarity Measurement User Preferences User Rating Analysis

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Review of Leap Motion Controller-Based Hand Gesture Datasets

A Comprehensive Review of Leap Motion Controller-Based Hand ...

引用

2023 IEEE International Conference on Next Generation Electronics, NEleX 2023

作者： Chakravarthi, Bharatesh Prabhu Prasad, B.M. Imandi, Raju Pavan Kumar, B.N. School of Computing and Augmented Intelligence Arizona State University TempeAZ United States Indian Institute of Information Technology Dept. of Computer Science and Engineering Dharwad India Indian Institute of Information Technology Sri City Dept. of Computer Science and Engineering Chittoor India

ISBN: (纸本)9798350319088

This paper comprehensively reviews hand gesture datasets based on Ultraleap's leap motion controller, a popular device for capturing and tracking hand gestures in real-time. The aim is to offer researchers and practitioners a valuable resource for developing and evaluating gesture recognition algorithms. The review compares various datasets found in the literature, considering factors such as target domain, dataset size, gesture diversity, subject numbers, and data modality. The strengths and limitations of each dataset are discussed, along with the applications and research areas in which they have been utilized. An experimental evaluation of the leap motion controller 2 device is conducted to assess its capabilities in generating gesture data for various applications, specifically focusing on touchless interactive systems and virtual reality. This review serves as a roadmap for researchers, aiding them in selecting appropriate datasets for their specific gesture recognition tasks and advancing the field of hand gesture recognition using leap motion controller technology. © 2023 IEEE.

关键词： Gesture recognition

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Meteorological Factors’ Effect on Daily Heart Rate Variability Over Five Years 12

Evaluation of Meteorological Factors’ Effect on Daily Heart...

引用

12th International Conference on Awareness science and Technology, iCAST 2023

作者： Kokubun, Kota Li, Tianhui Chen, Wenxi Dept. of Computer and Information Systems The University of Aizu Aizu-Wakamatsu Japan Division of Information System The University of Aizu Aizu-Wakamatsu Japan Dept. of Computer Science and Engineering The University of Aizu Aizu-Wakamatsu Japan

ISBN: (纸本)9798350324693

The aim of this study is to investigate the effect of meteorological factors on heart rate variability (HRV) derived from daily electrocardiogram (ECG) data. Daily meteorological data were obtained from the Japan Meteorological Agency’s database. Daily ECG data were measured while bathing for five y ears. T he s ubject w as a h ealthy m ale i n h is fi fties. After collecting ECG data, R-R intervals were calculated to perform HRV analysis in time domain and frequency domain. The evaluation of link between meteorological factors and HRV parameters was realized by utilizing meteorological factors’ importance calculated by Random Forest, and Steel-Dwass significant difference analysis of the values. Additionally, the evaluation was also performed by analyzing the obtained meteorological factors’ importance values using Granger Causality test. In Steel-Dwass test, several significant d ifferences (p © 2023 IEEE.

关键词： Electrocardiograms

来源：评论

学校读者我要写书评

暂无评论

Navigating the Privacy and Cybersecurity Risks of Smart Homes 24

Navigating the Privacy and Cybersecurity Risks of Smart Home...

引用

24th International Arab Conference on information Technology, ACIT 2023

作者： Alghamdi, Fatimah Abdulrahman Alghamdi, Abdulrahman Abdullah College of Computing and Artificial Intelligence Jeddah University Dept. Computer Science and Engineering Jeddah Saudi Arabia College of Computing and Information Technology Shaqra University Dept. Computer Science Shaqra Saudi Arabia

ISBN: (纸本)9798350384307

The increasing popularity of smart home devices has led to a growing concern over cyberthreats to these devices. This study aims to investigate the perceived cyberthreats to smart home devices and identify ways to protect privacy and countermeasure against these threats. Five experts in the field of smart home technology were selected for this study. Semi-structured interviews were conducted with the selected experts over the telephone. The transcripts of the interviews were analyzed using thematic analysis. The main findings of the study indicate that the most significant cyberthreats to smart home devices are: Voice recordings (35.5%), Camera and microphone access (30.1 %), Location tracking (24.9%). Personal information (20.1%), Data breaches (15.1 %), Third-party access (7.4 %), Privacy invasion (7.4%), Lack of control over personal information (5.0%), Cyber-attacks (4.8%), Security vulnerabilities in smart devices (4.8%) were also identified as potential threats. The study suggests that protecting personal data and ensuring device security are key considerations for safeguarding smart home devices and that it is important to take the necessary precautions to protect smart home devices from cyber threats, and to be aware of the potential risks that these devices pose to personal privacy and security. © 2023 IEEE.

关键词： Countermeasures Cyberthreats Data breaches IOT security Privacy invasion Smart home devices

来源：评论

学校读者我要写书评

暂无评论

Enhancing Power Grid Resilience: Advanced Strategies for Detecting and Analyzing Faults in Three-Phase Transmission Lines

Enhancing Power Grid Resilience: Advanced Strategies for Det...

引用

2024 IEEE International Conference on information Technology, Electronics and Intelligent Communication Systems, ICITEICS 2024

作者： Ahmed, Khandakar Rabbi Shayer, Khandokar Shadman Shuvo, Jahidul Islam Haque, F.M. Mahmudul Salam, Towfika Miraz, Shafiun Noman, Abdullah Al Sobuj, Md. Shohanur Islam Arafat, Md. Yaqub Rial, Md Istiak Hasan Miyan Research Institute International University of Business Agriculture and Technology Uttara Model Town Dhaka Bangladesh Dept. of Mechatronics Engineering Kazla Rajshahi6204 Bangladesh American International University-Bangladesh Dept. of Electrical & Electronic Engineering Dhaka Bangladesh Ahsanullah University of Science and Technology Dept. of Electrical & Electronic Engineering Dhaka Bangladesh Dept. of Computer Science and Engineering Rajshahi Bangladesh Jahangirnagar University Dept. of Institute of Information Technology Saver Dhaka1342 Bangladesh Hajee Mohammad Danesh Science and Technology University Dept. of Electrical and Electronic Engineering Dhaka Bangladesh International University of Business Agriculture and Technology Dept. of Mechanical Engineering Dhaka Bangladesh Dept. of Computer Science and Engineering Andhra Pradesh Kakinada533291 India

ISBN: (纸本)9798350382693

Power system fault analysis is a process that establishes the bus voltages and line currents when various fault types occur. Three-phase balanced faults and unbalanced faults are two categories of power system faults. Single line-to-ground, line-to-line, and double line-to-ground faults are three different types of unbalanced faults that can occur on transmission lines for power systems. Fault analysis is used to choose and set the proper protective devices and switchgear. Determining the bus voltages and line currents is critical while investigating power system issues. Numerous mathematical calculation techniques are shown in the process, which is challenging to do by hand. The calculation can be performed quickly on a computer using a Simulink created using MATLAB. However, an error occurs when inverting an extensive bus system's Y-bus to Z-bus in the conventional short circuit study;this paper offers a solution that is solved using 'MATLAB SIMULINK' to reduce error. © 2024 IEEE.

关键词： Power system analysis

来源：评论

学校读者我要写书评

暂无评论

Dengue Cases Prediction and Analysis using LSTM and Region-specific Data

Dengue Cases Prediction and Analysis using LSTM and Region-s...

引用

2024 OPJU International Technology Conference on Smart Computing for Innovation and Advancement in Industry 4.0, OTCON 2024

作者： Ud Deen Shafique, Hafiz Noor Ali, Nauman Hussain, Muhammad Zunnurain Hasan, Muhammad Zulkifl Mustafa, Muzzamil Khalid, Aqsa Awan, Rimsha Ashraf, Farhan Khan, Zohaib Ahmed Javaid, Arslan University Lahore Department of Computer Engineering Information Technology Punjab Pakistan Bahria University Lahore Campus Dept. of Computer Science Zunnurain Pakistan University of Central Punjab Lahore Faculty of Information Technology Department of Computer Science Pakistan National College of Business Administration and Economics Department of Computer Science Lahore Pakistan Information Technology University Lahore Pakistan

ISBN: (纸本)9798350373783

Dengue Fever presents a substantial global health challenge, necessitating the development of accurate predictive models for effective disease management. This research paper conducts a comprehensive analysis of a Long ShortTerm Memory (LSTM) neural network-based model for forecasting Dengue cases in the Philippines. The study includes data preprocessing, model development using TensorFlow and Keras, and thorough evaluation using metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), Mean Absolute Percentage Error (MAPE), and Rsquared (R2). The dataset, 'Dengue Cases in the Philippines,' provides a detailed historical record of Dengue fever cases and fatalities across various regions, enabling a thorough analysis of temporal and spatial dynamics. Data preprocessing involves label encoding, normalization, and careful data splitting to ensure accurate model training and testing. The LSTM neural network architecture is optimized through hyperparameter tuning, including learning rate, batch size, and epochs, to improve prediction accuracy. Model performance is rigorously evaluated, with an MAE of 913, indicating areas for further improvement. While the model shows promise, there is room for refinement. This study contributes to the field by exploring the feasibility of LSTM models for Dengue prediction, with insights that may inform future research in disease forecasting and public health strategies. Limitations, ethical considerations, and recommendations for further work are discussed. © 2024 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

没有更多数据了...

全选清除本页清除全部题录导出标记到“检索档案”

共500页 << < 25 26 27 28 29 30 31 32 33 34 > >>

检索报告对象比较合并检索0

隐藏清空

合并搜索

回到顶部

执行限定条件

内容：

评分：

请选择保存的检索档案：

请选择收藏分类：

订阅名称：

通借通还

温馨提示：

图书名称：

借书校区：

取书校区：

手机号码：

邮箱地址：

一卡通帐号：

电话和邮箱必须正确填写，我们会与您联系确认。

联系人：

所在院系：

联系邮箱：

联系电话：

内蒙古自治区呼和浩特市赛罕区大学西街235号邮编: 010021

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：