检索结果-内蒙古大学图书馆

IEEE International conference on Robotics and Automation (ICRA)

作者： Mingxing Peng Zhihao Gong Chen Sun Long Chen Dongpu Cao School of Data and Computer Science Sun Yat-sen University China Waterloo CogDrive Lab University of Waterloo Canada

ISBN: (数字)9781728173955

ISBN: (纸本)9781728173962

Autonomous urban driving navigation is still an open problem and has ample room for improvement in unknown complex environments and terrible weather conditions. In this paper, we propose a two-stage framework, called IPP-RL, to handle these problems. IPP means an Imitation learning method fusing visual information with the additional steering angle calculated by Pure-Pursuit (PP) method, and RL means using Reinforcement Learning for further training. In our IPP model, the visual information captured by camera can be compensated by the calculated steering angle, thus it could perform well under bad weather conditions. However, imitation learning performance is limited by the driving data severely. Thus we use a reinforcement learning method-Deep Deterministic Policy Gradient (DDPG)-in the second stage training, which shares the learned weights from pretrained IPP model. In this way, our IPP-RL can lower the dependency of imitation learning on demonstration data and solve the problem of low exploration efficiency caused by randomly initialized weights in reinforcement learning. Moreover, we design a more reasonable reward function and use the n-step return to update the critic-network in DDPG. Our experiments on CARLA driving benchmark demonstrate that our IPP-RL is robust to lousy weather conditions and shows remarkable generalization capability in unknown environments on navigation task.

关键词： Learning (artificial intelligence) Meteorology Robustness Task analysis Navigation Training Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Sentiment analysis as an Indicator to Evaluate Gender disparity on Sexual Violence Tweets in South Africa

Sentiment Analysis as an Indicator to Evaluate Gender dispar...

引用

Robotics and Mechatronics conference (RobMech)

作者： Jude Oyasor Mpho Raborife Pravesh Ranchod School of Computer Science and Applied Mathematics University of the Witwatersrand Johannesburg South Africa School of Consumer Intelligence and Information Systems University of Johannesburg Johannesburg South Africa

ISBN: (数字)9781728141626

ISBN: (纸本)9781728141633

South Africa has one of the highest rates of sexual violence in the world, and with this technology era being the information age, it is not suprising that people are turning to social media to voice their views on this matter. As tweets on sexual violence grow continually, extracting insights from such data demands a robust, real-time, and scalable tools with flexibility in the database schema. Elasticsearch (ES) is an example of a free-license search engine written in Java and developed on Apache Lucene that meets the stated requirements. On the other hand, Kibana facilitates intuitive dashboard development, visual exploration, and real-time analysis of an index in ES through an intuitive graphical user interface. This study demonstrates how gender was inferred and evaluated through the integration of deep neural networks and Google's TensorFlow. We used the AFINN model to infer sentiment analysis as our measure of gender disparity in this instance. The Indexer, built of Node. js, and defined as the hub of the system connects with the Twitter streaming API to ingest tweets found within the boundaries of sexual violence. This system runs persistently with tweets through the ES search engine and visualised in Kibana.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A blueprint for semantically lifting field trial data: Enabling exploration using knowledge graphs 12

A blueprint for semantically lifting field trial data: Enabl...

引用

12th International conference on Semantic Web Applications and Tools for Health Care and Life Sciences, SWAT4HCLS 2019

作者： Burkow, D. Hollunder, J. Heinrich, J. Abdallah, F. Rojas-Macias, M.A. Wiljes, C. Cimiano, P. Senger, P. Bayer CropScience CLS Translational R&D Germany Semalytix GmbH Bielefeld Germany Semantic Computing Group Bielefeld University Germany

In many organizations, relevant data is distributed across multiple primary data sources using different data models, schemas, and vocabularies which prevents effective knowledge management at an organizational level. As a result, it is very cumbersome to retrieve and integrate relevant data necessary to answer specific business questions. We propose a semantic lifting architecture that lifts primary data resources into a single corporate knowledge graph, implemented following FAIR and Linked data principles, which assigns unique URIs to entities and maps the data using a uniform shared vocabulary. This structure supports integration of heterogeneous and distributed data sources to enable advanced search and analysis. In this paper we consider the case of field trial data collected and managed by Bayer, present an example of our architecture, and describe an application using this architecture that provides semantic search and visual exploration of field trial data to relevant stakeholders. As a main benefit, our architecture allows domain experts and analysts to retrieve timely and consistently labeled data originating from multiple primary sources while maintaining data context to enable cross-domain analytics. This provides the means to answer more complex questions with minimal data preparation, improving speed and precision by driving decisions with semantically lifted data. Copyright © 2019 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Sequence Synopsis: Optimize visual Summary of Temporal Event data

引用

IEEE TRANSACTIONS ON visualIZATION AND COMPUTER GRAPHICS 2018年第1期24卷 45-55页

作者： Chen, Yuanzhe Xu, Panpan Ren, Liu Hong Kong Univ Sci & Technol Hong Kong Hong Kong Peoples R China Bosch Res North Amer Palo Alto CA USA

Event sequences analysis plays an important role in many application domains such as customer behavior analysis, electronic health record analysis and vehicle fault diagnosis. Real-world event sequence data is often noisy and complex with high event cardinality, making it a challenging task to construct concise yet comprehensive overviews for such data. In this paper. we propose a novel visualization technique based on the minimum description length (MDL) principle to construct a coarse-level overview of event sequence data while balancing the information loss in it. The method addresses a fundamental trade-off in visualization design: reducing visual clutter vs. increasing the information content in a visualization. The method enables simultaneous sequence clustering and pattern extraction and is highly tolerant to noises such as missing or additional events in the data. Based on this approach we propose a visual analytics framework with multiple levels-of-detail to facilitate interactive data exploration. We demonstrate the usability and effectiveness of our approach through case studies with two real-world datasets. One dataset showcases a new application domain for event sequence visualization, i.e., fault development path analysis in vehicles for predictive maintenance. We also discuss the strengths and limitations of the proposed method based on user feedback.

关键词： Time Series data data Transformation and Representation visual Knowledge Representation visual Analytics

来源：评论

学校读者我要写书评

暂无评论

Active Auditory exploration for Identifying Object Contents

Active Auditory Exploration for Identifying Object Contents

引用

International conference on Advanced Robotics and Mechatronics (ICARM)

作者： Xiang Dou Huaping Liu Xinying Xu Di Guo College of Electrical and Power Engineering Taiyuan University of Technology Taiyuan China Tsinghua University State Key Laboratory of Intelligent Technology and Systems Beijing National Research Center for Information Science and Technology Beijing China

ISBN: (数字)9781728164793

ISBN: (纸本)9781728164809

Intelligent robots need to acquire a target object from the same type of container filled with different kinds of objects. The actions selected by robots are critical to quickly and accurately recognizing the type of objects inside the container. In this paper, we propose an active auditory exploration method based on reinforcement learning, which enables the robot to actively explore the operational behavior of interest and establish the coupling relationship between perception and action to reduce the ambiguity of target recognition. The robot interacts with visually indistinguishable bottles by adopting multiple action behaviors to generate sound data from which the perceptual model learns to classify object contents. This is similar to human beings actively exploring and accumulating experience through sound in the environment where vision cannot be judged. In addition, we also use passive strategies to compare and analyze the impact of inhibition behavior on object recognition. Experimental results show that our method successfully learns effective class recognition strategies and actively selects actions to further improve recognition efficiency.

关键词： Object recognition Containers Robot sensing systems Learning (artificial intelligence) Task analysis visual perception

来源：评论

学校读者我要写书评

暂无评论

Facilitating and Managing Machine Learning and data analysis Tasks in Big data Environments using Web and Microservice Technologies 19

Facilitating and Managing Machine Learning and Data Analysis...

引用

Proceedings of the 11th International conference on Management of Digital EcoSystems

作者： Shadi Shahoud Sonja Gunnarsdottir Hatem Khalloof Clemens Duepmeier Veit Hagenmeyer Karlsruhe Institute of Technology IAI Karlsruhe Germany

ISBN: (纸本)9781450362382

Driven by the great advance of machine learning in a wide range of application areas, the need for developing machine learning frameworks effectively as well as easily usable by novices increased dramatically. Furthermore, building machine learning models in the context of big data environments still represents a great challenge. In the present paper, we tackle these challenges by introducing a new generic framework for efficiently facilitating the training, testing, managing, storing, and retrieving of machine learning models in the context of big data. The framework makes use of a powerful big data software stack and a microservice architecture for a fully manageable and highly scalable solution. A highly configurable user interface is introduced giving the user the ability to easily train, test, and manage machine learning models. Moreover, it automatically indexes models and allows flexible exploration of them in the visual interface. The performance of the new framework is evaluated on state-of-the-arts machine learning algorithms: it is shown that storing and retrieving machine learning models as well as a respective acceptable low overhead demonstrate an efficient approach to facilitate machine learning in big data environments.

关键词： Big data Web-based Applications data Analytic Microservice Machine Learning

来源：评论

学校读者我要写书评

暂无评论

Spanning the Boundaries of data visualization Work: An exploration of Functional Affordances and Disciplinary Values 1

引用

14th International conference on Information in Contemporary Society, iconference 2019

作者： Snyder, Jaime Shilton, Katie University of Washington SeattleWA98195 United States University of Maryland College ParkMD20742 United States

ISBN: (数字)9783030157425

ISBN: (纸本)9783030157418

Creating data visualizations requires diverse skills including computer programming, statistics, and graphic design. visualization practitioners, often formally trained in one but not all of these areas, increasingly face the challenge of reconciling, integrating and prioritizing competing disciplinary values, norms and priorities. To inform multidisciplinary visualization pedagogy, we analyze the negotiation of values in the rhetoric and affordances of two common tools for creating visual representations of data: R and Adobe Illustrator. Features of, and discourse around, these standard visualization tools illustrate both a convergence of values and priorities (clear, attractive, and communicative data-driven graphics) side-by-side with a retention of rhetorical divisions between disciplinary communities (statistical analysis in contrast to creative expression). We discuss implications for data-driven work and data science curricula within the current environment where data visualization practice is converging while values in rhetoric remain divided. © 2019, Springer Nature Switzerland AG.

关键词： data visualization

来源：评论

学校读者我要写书评

暂无评论

PolyVis: Cross-Device Framework for Collaborative visual data analysis

PolyVis: Cross-Device Framework for Collaborative Visual Dat...

引用

IEEE International conference on Systems, Man and Cybernetics

作者： Abeer Alsaiari Andrew Johnson Arthur Nishimoto Electronic Visualization Laboratory University of Illinois at Chicago Chicago IL USA

Multi device environments present new opportunities for collaborative visual data analysis and sense making by utilizing each device's strengths and capabilities. However, one of the associated challenges with visual data analysis in multi device environments is the sharing of visual components across devices. We present a framework developed on top of SAGE2 platform for cross-device collaborative visual data exploration. As part of our framework, we contribute the concept of rapid development and assembling of visualizations that can span multiple devices of different modalities. It provides the users with an environment for visualization compositions that delegate the rendering to the target device, allowing them to augment their large display workspace with portable devices for further exploration territories. Facilitated by its intuitive visualization composition pipeline, users with no programming skills such as data analysts can enhance their analytical scope with no coding barriers. We describe the framework, its implementation with a use case, and the rationale behind its design.

关键词： Collaboration data visualization data analysis visual analytics Task analysis Headphones

来源：评论

学校读者我要写书评

暂无评论

Multi-target Image Detection Using Pigeon-Inspired Optimization based Species Mechanism

Multi-target Image Detection Using Pigeon-Inspired Optimizat...

引用

Information Technology,Big data and Artificial Intelligence (ICIBA), International conference on IEEE International conference on

作者： Long Wang Jing Fan Xiao Wei Xincheng Cui Dongjie Wu National Computer System Engineering Research Institute of China Beijing China State-owned Changhong Machinery Factory Guilin China

ISBN: (数字)9781728152240

ISBN: (纸本)9781728152257

Multi-target detection (MTD) has a wide range of applications in target recognition, visual tracking, and other fields. The task of MTD is to find the sub-image set with the highest matching degree using the template image set from the source image. Therefore, this problem can be mapped to a multi-modal optimization problem. This paper combines the modified pigeon-inspired optimization (PIO) and species-based mechanism to perform multi-template matching tasks. To enhance the performance of our proposed method, the PIO is modified as follows: (1) the fitness of the seed is chosen as the switch signal of the map and compass operator and landmark operator; (2) in landmark operator, the halve mechanism is removed and a perturbation factor is introduced to enhance the exploration and exploitation ability. The species-based mechanism is utilized to search for multiple extreme points in parallel with multiple subpopulations. In addition, an expel mechanism is proposed to avoid the unnecessary fitness evaluation around the detected optima. Simulation results demonstrate that our proposed method performs better in detection accuracy and efficiency.

关键词： Optimization Statistics Sociology Compass Particle swarm optimization Switches Task analysis

来源：评论

学校读者我要写书评

暂无评论

CCVis: visual analytics of student online learning behaviors using course clickstream data

CCVis: Visual analytics of student online learning behaviors...

引用

2019 conference on visualization and data analysis, VDA 2019

作者： Goulden, Maggie Celeste Gronda, Eric Yang, Yurou Zhang, Zihang Tao, Jun Wang, Chaoli Duan, Xiaojing Ambrose, G. Alex Abbott, Kevin Miller, Patrick Trinity College Dublin Ireland University of Maryland Baltimore County United States Zhejiang University China University of Notre Dame United States

As more and more college classrooms utilize online plat- forms to facilitate teaching and learning activities, analyzing stu- dent online behaviors becomes increasingly important for instruc- tors to effectively monitor and manage student progress and per- formance. In this paper, we present CCVis, a visual analytics tool for analyzing the course clickstream data and exploring student online learning behaviors. Targeting a large college introductory course with over two thousand student enrollments, our goal is to investigate student behavior patterns and discover the possi- ble relationships between student clickstream behaviors and their course performance. We employ higher-order network and struc- tural identity classification to enable visual analytics of behavior patterns from the massive clickstream data. CCVis includes four coordinated views (the behavior pattern, behavior breakdown, clickstream comparative, and grade distribution views) for user interaction and exploration. We demonstrate the effectiveness of CCVis through case studies along with an ad-hoc expert evalua- tion. Finally, we discuss the limitation and extension of this work. © 2019, Society for Imaging Science and Technology.

关键词： Students

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：