Sensor Mining at work
2
pages
English
Documents
Le téléchargement nécessite un accès à la bibliothèque YouScribe Tout savoir sur nos offres
2
pages
English
Documents
Le téléchargement nécessite un accès à la bibliothèque YouScribe Tout savoir sur nos offres
Publié par
Langue
English
Sensor Mining at work: Principles and a Water QualityCase Study(SIGMOD06 Tutorial proposal)Christos Faloutsos Jeanne VanbriesenSchool of Computer Science, Department of Civil andCarnegie Mellon University Environmental EngineeringCarnegie Mellon Universitychristos@cs.cmu.edujeanne@cmu.edu1. INTENDED DURATION cation.3 hours3. INTENDED AUDIENCEResearchers and practitioners that want a concise, intu-2. MOTIVATION - BASIC INFORMATIONitive overview of the major tools in sensor mining, motivatedHow can we nd patterns in a collection of measurements,by the vital problem of water quality monitoring.say, on water quality sensors? Is the water safe to drink?Are we under biological attack? How many sensors do weneed to place, and where? 4. COVERAGEThe instructors have been collaborating on exactly theseproblems for the past 3 years. The tutorial will report our Problem de nition [Faloutsos]experiences. Speci cally, the tutorial surveys the related ar-eas and has two goals: (a) to review the main principles and Main tools [Faloutsos]main data base tools for sensor data analysis (b) to show-{ Time series and Forecastingcase them on a real, important application, namely drinkingwater quality. Time series indexing and feature extractionThe rst part will examine the state of the art in time se- Fourier, Wavelets, Time Warpingries indexing and mining. We will cover feature extraction, Linear forecasting, ARIMA, recursive leastpowerful tools from signal ...
Voir
Publié par
Langue
English
Sensor Mining at work:Principles and a Water Quality CaseStudy
(SIGMOD06 Tutorial proposal)
Christos Faloutsos School of Computer Science, Carnegie Mellon University christos@cs.cmu.edu
Jeanne Vanbriesen Department of Civil and Environmental Engineering Carnegie Mellon University
jeanne@cmu.edu
1. INTENDEDDURATIONcation. 3 hours 3. INTENDEDAUDIENCE 2. MOTIVATION BASIC INFORMATIONResearchers and practitioners that want a concise, intu-itive overview of the major tools in sensor mining, motivated How can we find patterns in a collection of measurements, by the vital problem of water quality monitoring. say, on water quality sensors?Is the water safe to drink? Are we under biological attack?How many sensors do we 4. COVERAGE need to place, and where? The instructors have been collaborating on exactly these problems for the past 3 years.The tutorial will report our •Problem definition [Faloutsos] experiences. Specifically,the tutorial surveys the related ar-eas and has two goals:(a) to review the main principles and•Main tools [Faloutsos] main data base tools for sensor data analysis (b) to show-–Time series and Forecasting case them on a real, important application, namely drinking water quality. ∗Time series indexing and feature extraction The first part will examine the state of the art in time se-∗Fourier, Wavelets, Time Warping ries indexing and mining.We will cover feature extraction, ∗Linear forecasting, ARIMA, recursive least powerful tools from signal processing (Fourier, Wavelets), squares and traditional methods for mining and forecasting:the Box-Jenkins (AutoRegressive) methodology.We will also–Lag correlations cover powerful methods for discovering correlations across –Cross correlations co-evolving time sequences, like Singular Value Decomposi-∗Hidden variable detection tion (SVD) and Blind Source Separation (BSS), also known as Independent Component Analysis (ICA).∗Singular value decomposition The second part will review the state of the art of water ∗Sketches and random projections sensors, their capabilities and limitations, the correspond-∗Independent Component Analysis ing research challenges they induce to the sensor mining algorithms. Itwill also describe the problem of monitoring •Intro to water quality [Vanbriesen] multiple substances that chemically react with each other, and how to infer concentrations of one substance, given mea-–Types of sensors now and in the future surements of the other.Finally, it will describe the ’control’ –Direct and surrogate data collection problem of where to optimally place a fixed number of sen-–Integration of sensors into control systems sors, to maximize the monitoring benefits. The tutorial ends with a list of future directions for database –Modeling Distribution systems with and without research, motivated by the water quality monitoring appli-potential sensors –Recent achievements in multi-species modeling mod-eling Permission to make digital or hard copies of all or part of this work for–Future directions in sensor data for water sys-personal or classroom use is granted without fee provided that copies are tems. not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to •Conclusions [Faloutsos] republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. –Database research directions for water quality mon-ACM/SIGMOD’06 Chicago, Illinois USA itoring. Copyright 2006 ACM XXXXXXXXX/XX/XX ...$5.00.
1
5. SPECIALEQUIPMENT None 6. OTHERINFORMATION •The tutorial is a spin-off of a large NSF joint project on water quality monitoring, which we reported in [1]. •The first part of the tutorial, has overlap with older tutorials by Faloutsos.The differences are (a) coverage of additional methods, like Blind Source Separation, lag correlation discovery, and on-line SVD, and (b) emphasis on water-quality applications. 7. BIOGRAPHICALNOTES Christos Faloutsosis a Professor at Carnegie Mellon Uni-versity. Hehas received the Presidential Young Investigator Award by the National Science Foundation (1989), six “best paper” awards, and several teaching awards.He served as a member of the executive committee of SIGKDD; he has published over 130 refereed articles, one monograph, and holds five patents.His research interests include data min-ing for streams, sensors and networks, fractals, indexing for multimedia and bio-informatics data bases, and database performance. Jeanne Vanbriesenis an Associate Professor at Carnegie Mellon University in the Department of Civil and Envi-ronmental Engineering.She has received the George Tall-mann Ladd Award for outstanding research and professional accomplishments from the College of Engineering and is the first recipient of the Paul and Norene Christiano Fac-ulty Fellowship.Her research interests include detection of pathogens in water, applications of thermodynamics to bac-terial systems, and biodegradation of recalcitrant organic compounds. 8. REFERENCES [1] A.Ailamaki, C. Faloutsos, P. S. Fischbeck, M. J. Small, and J. VanBriesen. An environmental sensor network to determine drinking water quality and security. SIGMOD Record, 32(4):47–52, 2003.
2
