0


挖掘大量RFID、轨迹和交通数据集

Mining Massive RFID, Trajectory, and Traffic Data Sets
课程网址: http://videolectures.net/kdd08_han_mmrfid/  
主讲教师: Jiawei Han; Jae-Gil Lee; Hector Gonzalez; Xiaolei Li
开课单位: 视频讲座网
开课时间: 2008-09-26
课程语种: 英语
中文简介:
随着卫星、RFID、GPS、传感器、无线和视频技术的广泛应用,移动目标数据已被大规模收集,并变得越来越丰富、复杂和无处不在。对移动目标信息进行可扩展和灵活的数据分析的需求迫在眉睫;因此,对移动对象数据的挖掘已成为数据挖掘的主要挑战之一。对于RFID、轨迹和交通数据集的数据挖掘,已经进行了大量的研究。然而,目前还没有从这种移动对象数据集中发现知识的系统教程。本教程对分析不同类型移动对象数据集的方法和算法进行了全面、有组织和最新的调查,并强调了几个重要的挖掘任务:聚类、分类、离群值分析和多维分析。除了对这一主题的近期研究工作的全面调查之外,我们还展示了如何从RFID、轨迹和交通数据集的数据挖掘中获益。本教程由三部分组成:(1)RFID数据挖掘,(2)轨迹数据挖掘,(3)交通数据挖掘。在第一部分中,探讨了RFID数据的仓库、清洁和流挖掘。第二部分研究了弹道数据的模式挖掘、聚类、分类和离群点检测。第三部分探讨了交通数据的路由发现、目的地预测和热点路由或离群点检测。本教程是为对移动对象数据感兴趣的数据挖掘、数据库和机器学习研究人员编写的。
课程简介: With the wide availability of satellite, RFID, GPS, sensor, wireless, and video technologies, moving-object data has been collected in massive scale and is becoming increasingly rich, complex, and ubiquitous. There is an imminent need for scalable and flexible data analysis over moving-object information; and thus mining moving-object data has become one of major challenges in data mining. There have been considerable research efforts on data mining for RFID, trajectory, and traffic data sets. However, there has been no systematic tutorial on knowledge discovery from such moving-object data sets. This tutorial presents a comprehensive, organized, and state-of-the-art survey on methodologies and algorithms on analyzing different kinds of moving-object data sets, with an emphasis on several important mining tasks: clustering, classification, outlier analysis, and multidimensional analysis. Besides a thorough survey of the recent research work on this topic, we also show how real-world applications can benefit from data mining of RFID, trajectory, and traffic data sets. The tutorial consists of three parts: (1) RFID data mining, (2) trajectory data mining, and (3) traffic data mining. In the first part, warehousing, cleaning, and flow mining for RFID data are explored. In the second part, pattern mining, clustering, classification, and outlier detection for trajectory data are explored. In the third part, route discovery, destination prediction, and hot-route or outlier detection for traffic data are explored. This tutorial is prepared for data mining, database, and machine learning researchers who are interested in moving-object data.
关 键 词: 移动数据; 数据分析; 数据挖掘; 移动对象
课程来源: 视频讲座网
数据采集: 2022-11-30:chenxin01
最后编审: 2022-11-30:chenxin01
阅读次数: 23