U.S.-Japan Joint Seminar: A Data Mining-Based Approach to Semantic Integration of Stream Data

美日联合研讨会：基于数据挖掘的流数据语义整合方法

• 批准号: 0318547
• 负责人: Christos Faloutsos
• 金额: $ 1.58万
• 依托单位: Carnegie-Mellon University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-09-01 至 2005-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0318547&HistoricalAwards=false
• 关键词: U.S.; Japan; Joint; Seminar; Data

0318547FaloutsosThis award supports a two-year international collaborative research project. The principal investigator for the project is Professor Christos Faloutsos at Carnegie Mellon University in Pittsburgh, Pennsylvania and the Japanese collaborator is Hiroyuki Kitagawa at the University of Tsukuba in Japan. They will be undertaking research on a data mining-based approach to semantic integration of stream data. Recent progress of high-speed network, mobile communication, and low-cost sensor device technologies permit the continuous collection of stream data such as temperatures, stock prices and news feeds. Stream data can be acquired from various sources, for example, sensors, cameras and information dissemination servers such as news delivery channels. The goal of the research is to find patterns and outliers and strange behavior, such as - these temperatures are abnormally high given the respective humidities. One of the problems in stream data is event detection. They hope to develop data mining techniques to extract events and metadata automatically and efficiently, by analyzing the given streams. They also propose to develop event-driven data integration techniques to take appropriate actions (e.g., issue alerts) according to the events detected. They further hope to develop stream classification methods and data integration methods to merge stream information with traditional information sources such as relational databases. Finally they plan to implement all of the above tools in a prototype system with a visual interface. The project brings together the efforts of two groups that have complementary expertise and research capabilities. These joint efforts should have a positive impact in many areas such as (1) civil infrastructure - sensors on bridges and highways measuring wear and tear, number of cars and road conditions; (2) military applications - with smart cheap sensors measuring ground and air conditions; (3) environmental applications - with sensors measuring water and air pollution and contamination levels. The project advances international human resources through the participation of graduate students. Through the exchange of ideas and technology, this project will broaden our base of basic knowledge and promote international understanding and cooperation. Results of the research will be disseminated at scientific meetings and in scientific journals.

0318547 Faloutsos该奖项支持一个为期两年的国际合作研究项目。 该项目的主要研究者是宾夕法尼亚州匹兹堡卡内基梅隆大学的Christine Faloutsos教授，日本合作者是日本筑波大学的Hiroyuki Kitagawa。 他们将研究基于数据挖掘的流数据语义集成方法。 高速网络、移动的通信和低成本传感器设备技术的最新进展允许连续收集诸如温度、股票价格和新闻馈送的流数据。 流数据可以从各种源获取，例如传感器、摄像机和信息传播服务器（例如新闻传送频道）。 研究的目标是找到模式和异常值和奇怪的行为，例如-这些温度在各自的湿度下异常高。 流数据中的问题之一是事件检测。 他们希望开发数据挖掘技术，通过分析给定的流来自动有效地提取事件和元数据。 他们还建议开发事件驱动的数据集成技术，以采取适当的行动（例如，发出警报）。 他们进一步希望开发流分类方法和数据集成方法，将流信息与关系数据库等传统信息源合并。 最后，他们计划在一个具有可视化界面的原型系统中实现上述所有工具。 该项目汇集了两个具有互补专业知识和研究能力的团体的努力。 这些共同努力应该在许多领域产生积极影响，例如（1）民用基础设施-桥梁和高速公路上的传感器测量磨损，汽车数量和道路状况;（2）军事应用-智能廉价传感器测量地面和空气状况;（3）环境应用-传感器测量水和空气污染和污染水平。 该项目通过研究生的参与促进国际人力资源。 通过交流思想和技术，该项目将扩大我们的基础知识，促进国际理解与合作。 研究结果将在科学会议和科学期刊上传播。