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

4,579 篇 会议
572 篇 期刊文献
10 篇 学位论文
5 册 图书
1 篇 科技报告

馆藏范围

5,167 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

3,667 篇 工学
- 2,224 篇 计算机科学与技术...
- 1,441 篇 软件工程
- 576 篇 电气工程
- 380 篇 控制科学与工程
- 331 篇 信息与通信工程
- 258 篇 机械工程
- 249 篇 石油与天然气工程
- 214 篇 化学工程与技术
- 204 篇 生物医学工程（可授...
- 194 篇 生物工程
- 189 篇 仪器科学与技术
- 172 篇 土木工程
- 163 篇 光学工程
- 154 篇 建筑学
- 151 篇 动力工程及工程热...
- 132 篇 材料科学与工程（可...
- 124 篇 电子科学与技术（可...
- 121 篇 航空宇航科学与技...
1,585 篇 理学
- 600 篇 物理学
- 418 篇 数学
- 313 篇 地球物理学
- 270 篇 地质学
- 220 篇 生物学
- 202 篇 统计学（可授理学、...
- 191 篇 化学
- 120 篇 系统科学
656 篇 管理学
- 419 篇 管理科学与工程(可...
- 265 篇 图书情报与档案管...
- 172 篇 工商管理
400 篇 医学
- 307 篇 临床医学
90 篇 法学
86 篇 农学
67 篇 教育学
59 篇 经济学
30 篇 文学
12 篇 艺术学
4 篇 军事学
1 篇 历史学

主题

664 篇 visualization
381 篇 data visualizati...
243 篇 visual analytics
170 篇 data mining
149 篇 data analysis
135 篇 feature extracti...
101 篇 data models
94 篇 deep learning
92 篇 image color anal...
90 篇 task analysis
88 篇 training
81 篇 accuracy
79 篇 computational mo...
72 篇 analytical model...
71 篇 big data
66 篇 machine learning
62 篇 cameras
57 篇 predictive model...
57 篇 semantics
56 篇 convolutional ne...

机构

17 篇 center for space...
17 篇 nagoya universit...
17 篇 department of el...
16 篇 infn sezione di ...
16 篇 department of ph...
16 篇 quantum ict adva...
16 篇 department of ph...
16 篇 department of ph...
15 篇 vrvis res ctr au...
15 篇 institute of spa...
15 篇 faculty of scien...
15 篇 infn sezione di ...
15 篇 institute of par...
15 篇 national institu...
15 篇 via madonna del ...
15 篇 department of as...
15 篇 faculty of engin...
14 篇 univ konstanz co...
14 篇 waseda research ...
14 篇 jem utilization ...

作者

22 篇 asano k.
18 篇 schreck tobias
18 篇 cherry m.l.
18 篇 krizmanic j.f.
17 篇 tamura t.
17 篇 bruno a.
17 篇 kawakubo y.
17 篇 katayose y.
17 篇 binns w.r.
17 篇 castellini g.
17 篇 israel m.h.
17 篇 uchihori y.
17 篇 ebisawa k.
17 篇 yoshida a.
17 篇 bongi m.
17 篇 bigongiari g.
17 篇 shiomi a.
17 篇 collazuol g.
17 篇 nakahira s.
17 篇 ozawa s.

语言

4,705 篇 英文
253 篇 其他
161 篇 俄文
48 篇 中文
3 篇 乌克兰文
2 篇 德文
1 篇 法文
1 篇 葡萄牙文

检索条件"任意字段=Conference on Visual Data Exploration and Analysis V"

共 5167 条记录，以下是501-510 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：

9th International conference on Mining Intelligence and Knowledge exploration, MIKE 2023

9th International Conference on Mining Intelligence and Know...

引用

9th International conference on Mining Intelligence and Knowledge exploration, MIKE 2023

ISBN: (纸本)9783031440830

The proceedings contain 38 papers. The special focus in this conference is on International conference on Mining Intelligence and Knowledge exploration. The topics include: Highest Accuracy Based Automated Depression Prediction Using Natural Language Processing;crime Prediction Using Modified Capsule Network with CrissCross Optimization on the Sentiment analysis for Cyber-Security;sentiment analysis Using Lexical Approach and Fuzzy Logic;HTTP, WebSocket, and SignalR: A Comparison of Real-Time Online Communication Protocols;an IoT Based Early Alert System to Monitor and Reduce Electrical Energy Consumption at Home in Smart Cities;Securing the MANET by Detecting the Flooding Attacks Using Hybrid CNN-Bi-LSTM-RF Model;The Smart Coverage Path Planner for Autonomous Drones Using TSP and Tree Selection;UAv Smart Navigation: Combining Delaunay Triangulation and the Bat Algorithm for Enhanced Efficiency;a Comparative analysis of data Backup and Network Consistency in Cluster-Base Wireless Sensor Network Protocols;development of IoT-Healthcare Model for Gastric Cancer from Pathological Images;Detection of Chicken Disease Based on Day-Age Using Pre Trained Model of CNN;Glaucoma Detection Using the YOLO v5 Algorithm;Deep Learning Ocular Disease Detection System (ODDS);efficient Chest X-Ray Investigation Using Firefly Algorithm Optimized Deep and Handcrafted Features;disease Detection and Risk Prediction System Based Web Application Using Machine Learning;Weighted Average Ensemble Approach for Pediatric Pneumonia Diagnosis Using Channel Attention Deep CNN Architectures;seasonal Disease Based Demand Forecasting for Pharmaceutical Medications Using Random Forest;hybrid Optimal Fine Tuning Approach in Deep Learning for Identifying Early Parkinson’s Disease;cryptocurrency Price Prediction Using Deep Learning;ensemble Learning Based Social Engineering Fraud Detection Module for Cryptocurrency Transactions.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Grounding Partially-Defined Events in Multimodal data

Grounding Partially-Defined Events in Multimodal Data

引用

2024 conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Sanders, Kate Kriz, Reno Etter, David Recknor, Hannah Martin, Alexander Carpenter, Cameron Lin, Jingyang van Durme, Benjamin Johns Hopkins University United States University of Rochester United States

ISBN: (纸本)9798891761681

How are we able to learn about complex current events just from short snippets of video? While natural language enables straightforward ways to represent under-specified, partially observable events, visual data does not facilitate analogous methods and, consequently, introduces unique challenges in event understanding. With the growing prevalence of vision-capable AI agents, these systems must be able to model events from collections of unstructured video data. To tackle robust event modeling in multimodal settings, we introduce a multimodal formulation for partially-defined events and cast the extraction of these events as a three-stage span retrieval task. We propose a corresponding benchmark for this task, MultivENT-G, that consists of 14.5 hours of densely annotated current event videos and 1,168 text documents, containing 22.8K labeled event-centric entities. We propose a collection of LLM-driven approaches to the task of multimodal event analysis, and evaluate them on MultivENT-G. Results illustrate the challenges that abstract event understanding poses and demonstrates promise in event-centric video-language systems. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Design of Secure IoMT Networks Using a Federated Learning Approach 5th

Design of Secure IoMT Networks Using a Federated Learning Ap...

引用

5th International conference on Advances in Information Communication Technology and Computing, AICTC 2024

作者： Sahu, Hemant Joshi, N.K. Chande, Swati v. Rajasthan Technical University Kota India Modi Institute of Management and Technology Kota India Department of MCA International School of Informatics and Management Technical Campus Jaipur India

ISBN: (纸本)9789819761050

The Internet of Multimedia Things (IoMT) is gaining speed due to the increase in multimedia data and the interconnectedness of real-world physical devices with the advent of the Internet of Things (IoT) and real-time multimedia data transmission services. The paper focuses on the IoMT's unique challenges and potential, particularly in the domains of security, privacy, and network efficiency. Our research pivots around two core areas: addressing network vulnerabilities in IoMT and enhancing data privacy, specifically in healthcare contexts. This includes an exploration of federated learning combined with homomorphic encryption, underlining their importance in maintaining data security and integrity. We investigate the balance between model performance and execution time, with a detailed analysis of classification metrics in both encrypted and plain data scenarios. This approach highlights the effectiveness of federated learning as a privacy-preserving method in healthcare IoMT applications. The synthesis of these findings illuminates the intricate balance required between security, privacy, and performance in IoMT. It underscores the potential of advanced ML models and privacy-preserving techniques in building a secure, resilient, and ethical digital ecosystem. Our study not only contributes to the advancements in network security and data privacy but also establishes a foundational framework for the responsible and innovative evolution of IoMT. This research fills a critical gap by focusing on the specific requirements and challenges of multimedia content in IoMT, advancing beyond the traditional IoT paradigms that predominantly consider scalar data. The IoMT model introduced here serves as a novel paradigm, envisioning smart, heterogeneous multimedia things interacting and cooperating to facilitate globally available multimedia-based services and applications. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Rich Screen Reader Experiences for Accessible data visualization

Rich Screen Reader Experiences for Accessible Data Visualiza...

引用

24th Eurographics/IEEE vGTC conference on visualization (Eurovis)

作者： Zong, Jonathan Lee, Crystal Lundgard, Alan Jang, JiWoong Hajas, Daniel Satyanarayan, Arvind MIT Cambridge MA 02139 USA Carnegie Mellon Univ Pittsburgh PA 15213 USA UCL London England

Current web accessibility guidelines ask visualization designers to support screen readers via basic non-visual alternatives like textual descriptions and access to raw data tables. But charts do more than summarize data or reproduce tables;they afford interactive data exploration at varying levels of granularity-from fine-grained datum-by-datum reading to skimming and surfacing high-level trends. In response to the lack of comparable non-visual affordances, we present a set of rich screen reader experiences for accessible data visualization and exploration. Through an iterative co-design process, we identify three key design dimensions for expressive screen reader accessibility: structure, or how chart entities should be organized for a screen reader to traverse;navigation, or the structural, spatial, and targeted operations a user might perform to step through the structure;and, description, or the semantic content, composition, and verbosity of the screen reader's narration. We operationalize these dimensions to prototype screen-reader-accessible visualizations that cover a diverse range of chart types and combinations of our design dimensions. We evaluate a subset of these prototypes in a mixed-methods study with 13 blind and visually impaired readers. Our findings demonstrate that these designs help users conceptualize data spatially, selectively attend to data of interest at different levels of granularity, and experience control and agency over their data analysis process. An accessible HTML version of this paper is available at: http://***/pubs/rich-screen-reader-vis-experiences.

关键词： data visualization

来源：评论

学校读者我要写书评

暂无评论

ImageNet-Architect Model for Accurate Cloud Sky and Sun Image Segmentation 5

ImageNet-Architect Model for Accurate Cloud Sky and Sun Imag...

引用

5th IEEE Global conference for Advancement in Technology, GCAT 2024

作者： Kalpana, S. Kanimozhi, K.v. Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350376685

Cloud, sky, and sun segmentation is a crucial task for various applications, including weather analysis, environmental monitoring, and image enhancement. Motivated by the need for accurate and efficient segmentation methods, the proposed ImageNet-Architect model, a robust encoder-decoder architecture designed to segment images into three distinct classes: clouds, sky, and sun. The aim of this study is to establish a reliable segmentation standard that can accurately classify each pixel in an image into the respective categories, thereby facilitating more precise analysis and interpretation of sky conditions. The model employs convolutional layers, batch normalization, ReLU activations, and skip connections to capture high-resolution features and maintain spatial information. The ImageNet-Architect model on a dataset comprising 512 images and their corresponding ground truth masks. The model was trained using data augmentation techniques to enhance its robustness to varying conditions. The results demonstrate the model's high accuracy, with Intersection over Union (IoU) scores of 0.88 for clouds, 0.91 for sky, and 0.87 for sun. visual comparisons between the segmented outputs and ground truth masks indicate that the model can effectively delineate the boundaries of clouds, sky, and sun under diverse conditions. The total parameters of the model are 31, 055, 297, of which 31, 043, 521 are trainable, reflecting the model's capacity for detailed feature learning. © 2024 IEEE.

关键词： Environmental monitoring

来源：评论

学校读者我要写书评

暂无评论

WHEN IS THE CLEANING OF SUBJECTIvE data RELEvANT TO TRAIN UGC vIDEO QUALITY METRICS? 29

WHEN IS THE CLEANING OF SUBJECTIVE DATA RELEVANT TO TRAIN UG...

引用

IEEE International conference on Image Processing (ICIP)

作者： Perrin, Anne-Flore Dormeval, Charles Wang, Yilin Birkbeck, Neil Adsumilli, Balu Le Callet, Patrick Nantes Univ Ecole Cent Nantes CAPACITES SAS CNRSLS2NUMR 6004 F-44000 Nantes France Google Inc Mountain View CA USA

ISBN: (数字)9781665496209

ISBN: (纸本)9781665496209

Outlier analysis and spammer detection recently gained momentum in order to reduce uncertainty of subjective ratings in image & video quality assessment tasks. The large proportion of unreliable ratings from online crowdsourcing experiments and the need for qualitative and quantitative large-scale studies in the deep-learning ecosystem played a role in this event. We study the effect that data cleaning has on trainable models predicting the visual quality for videos, and present results demonstrating when cleaning is necessary to reach higher efficiency. To this end, we present and analyze a benchmark on clean and noisy User Generated Content (UGC) large-scale datasets on which we re-trained models, followed by an empirical exploration of the constraint of data removal. Our results show that a dataset presenting between 7 and 30% of outliers benefits from cleaning before training.

关键词： Outlier Cleaning video Quality Metrics User Generated Content Training metrics

来源：评论

学校读者我要写书评

暂无评论

visual Object Tracking with Adaptive Template Update and Global Search Augmentation 3rd

Visual Object Tracking with Adaptive Template Update and Glo...

引用

3rd China Intelligent Robotics Annual conference, CCF CIRAC 2022

作者： Zeng, Lu He, Wei Zhang, Wenqiang Fudan University Shanghai China

ISBN: (纸本)9789819903009

The realization of human-machine-environment intimate interaction by intelligent robots is the research direction of cutting-edge exploration in the field of robotics. One of the important tasks is to realize active target tracking on the robot platform. Single-target tracking is subject to data changes such as target position and size in the video sequence, and is prone to target drift or loss when the environment changes drastically or is occluded. This paper aims at the application background of the intelligent foot robot platform, where deep learning technology is used to adopt adaptive multi-target tracking. The frame detection template updated Shuffle net v2–0.5 convolutional neural network builds a deep tracking model, which speeds up the model calculation. At the same time, the multi-template input ensures that the required target information can be located in a larger search image and a global search module is added. The target position re-detection is carried out, and the background enhancement training is integrated to significantly strengthen the discrimination ability of the global search network. The target tracking accuracy of the improved visual object tracking algorithm reaches 64.7%, and the accuracy of the target located at the center point of the marker frame reaches 86.8%, which is significantly improved compared with the traditional algorithm. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Target tracking

来源：评论

学校读者我要写书评

暂无评论

24th International conference on Artificial Intelligence in Education, AIED 2023

24th International Conference on Artificial Intelligence in ...

引用

24th International conference on Artificial Intelligence in Education, AIED 2023

ISBN: (纸本)9783031362712

The proceedings contain 79 papers. The special focus in this conference is on Artificial Intelligence in Education. The topics include: The Road Not Taken: Preempting Dropout in MOOCs;does Informativeness Matter? Active Learning for Educational Dialogue Act Classification;can virtual Agents Scale Up Mentoring?: Insights from College Students’ Experiences Using the *** Platform at an American Hispanic-Serving Institution;Real-Time AI-Driven Assessment and Scaffolding that Improves Students’ Mathematical Modeling during Science Investigations;Improving Automated Evaluation of Student Text Responses Using GPT-3.5 for Text data Augmentation;the Automated Model of Comprehension version 3.0: Paying Attention to Context;smartPhone: Exploring Keyword Mnemonic with Auto-generated verbal and visual Cues;analysing verbal Communication in Embodied Team Learning Using Multimodal data and Ordered Network analysis;improving Adaptive Learning Models Using Prosodic Speech Features;neural Automated Essay Scoring Considering Logical Structure;"Why My Essay Received a 4?": A Natural Language Processing Based Argumentative Essay Structure analysis;leveraging Deep Reinforcement Learning for Metacognitive Interventions Across Intelligent Tutoring Systems;enhancing Stealth Assessment in Collaborative Game-Based Learning with Multi-task Learning;how Peers Communicate Without Words-An Exploratory Study of Hand Movements in Collaborative Learning Using Computer-vision-Based Body Recognition Techniques;scalable Educational Question Generation with Pre-trained Language Models;involving Teachers in the data-Driven Improvement of Intelligent Tutors: A Prototyping Study;reflexive Expressions: Towards the analysis of Reflexive Capability from Reflective Text;Implementing and Evaluating ASSISTments Online Math Homework Support At large Scale over Two Years: Findings and Lessons Learned;algebra Error Classification with Large Language Models;exploration of Annotation Strategies for Automati

关键词：

来源：评论

学校读者我要写书评

暂无评论

Make Acoustic and visual Cues Matter: CH-SIMS v2.0 dataset and Av-Mixup Consistent Module 22

Make Acoustic and Visual Cues Matter: CH-SIMS v2.0 Dataset a...

引用

24th ACM International conference on Multimodal Interaction (ICMI)

作者： Liu, Yihe Yuan, Ziqi Mao, Huisheng Liang, Zhiyun Yang, Wanqiuyue Qiu, Yuanzhe Cheng, Tie Li, Xiaoteng Xu, Hua Gao, Kai Tsinghua Univ State Key Lab Intelligent Technol & Syst Dept Comp Sci & Technol Beijing Peoples R China Hebei Univ Sci & Technol Sch Informat Sci & Engn Shijiazhuang Hebei Peoples R China Beijing Natl Res Ctr Informat Sci & Technol BNRis Beijing Peoples R China China Agr Univ Beijing Peoples R China Beijing Univ Posts & Telecommun Beijing Peoples R China

ISBN: (纸本)9781450393904

Multimodal sentiment analysis (MSA), which supposes to improve text-based sentiment analysis with associated acoustic and visual modalities, is an emerging research area due to its potential applications in Human-Computer Interaction (HCI). However, existing researches observe that the acoustic and visual modalities contribute much less than the textual modality, termed as text-predominant. Under such circumstances, in this work, we emphasize making non-verbal cues matter for the MSA task. Firstly, from the resource perspective, we present the CH-SIMS v2.0 dataset, an extension and enhancement of the CH-SIMS. Compared with the original dataset, the CH-SIMS v2.0 doubles its size with another 2121 refined video segments containing both unimodal and multimodal annotations and collects 10161 unlabelled raw video segments with rich acoustic and visual emotion-bearing context to highlight non-verbal cues for sentiment prediction. Secondly, from the model perspective, benefiting from the unimodal annotations and the unsupervised data in the CH-SIMS v2.0, the Acoustic visual Mixup Consistent (Av-MC) framework is proposed. The designed modality mixup module can be regarded as an augmentation, which mixes the acoustic and visual modalities from different videos. Through drawing unobserved multimodal context along with the text, the model can learn to be aware of different non-verbal contexts for sentiment prediction. Our evaluations demonstrate that both CH-SIMS v2.0 and Av-MC framework enable further research for discovering emotion-bearing acoustic and visual cues and pave the path to interpretable end-toend HCI applications for real-world scenarios. The full dataset and code are available for use at https://***/thuiar/ch- sims- v2.

关键词： multimodal sentiment analysis dataset semi-supervised machine learning modality mixup

来源：评论

学校读者我要写书评

暂无评论

Evaluation of the void space structure and reservoir properties of rocks in the visevian and turnaisian formations of the Berezivske field of the Dnipro-Donetsk depression 17

Evaluation of the void space structure and reservoir propert...

引用

17th International conference Monitoring of Geological Processes and Ecological Condition of the Environment, Monitoring 2023

作者： Yemets, v.Y. Bezrodna, I.M. Taras Shevchenko National University of Kyiv Ukraine

ISBN: (纸本)9789462824874

Estimating the void space structure of reservoir rocks from well logs plays a significant role in the exploration and development of complex resources, but faces challenges. What's more, the methodology for studying the void space structure of complex reservoir rocks is paid little attention at present. In this paper, the typical visevian and turnaisian formations in the interval of 5800-6110 m in a well 203 of the Berezivske field of the Dnipro-Donetsk depression are used as an example. The data on compressional and shear wave velocities, as well as density, were utilized for a quantitative assessment of the void space structure. Based on the analysis of input data, the authors have established the most optimal initial approximations for the void formats of the investigated intervals as follows: intergranular voids - 0.699;transitional voids - 0.060 - 0.084;microcracks - 0.0028 - 0.0031;caverns - 4. Inversion of acoustic data for the characterization of the void structure was conducted using special software. Concentrations of void formats were identified. The calculated data revealed the contributions of different types of porosity to the total porosity of rock. It has been established that granular-cavernous porosity and fractured-cavernous-granular porosity are prevail for the visean and tournaisian deposits in a well 203 of the Berezivske field. © 2023 17th International conference Monitoring of Geological Processes and Ecological Condition of the Environment, Monitoring 2023. All rights reserved.

关键词： Porosity

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 << < 47 48 49 50 51 52 53 54 55 56 > >>

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：