x
WEB DATA evolved over time


                                          Real-Time Sensor,
                                 2010’s
                                          Social, Multi-media
                                                  data




                                            Dynamic User
                                 2000’s   Generated Content




                                 1990’s     Static Document
                                                and files


                             2
x
Properties of Streaming Data



              Huge Volume
                                       Rapid
   Continuous
                 Information
                  Overload!!
                                   Heterogeneous

                               3
x
Some Statistics


   “A cross-country flight from New York to Los
“Sensors Networksbeenproduce in the last the amount
     “More on a Boeing 737 plane 10-20 times massive
                     will created
   Angeles data has                generates a
of generated -bythan in mediasummarize this data”
     Solution “Meaningfully in the next few years”
     three years social all the past 40,000 years”
   240 terabytes of data”
 - GigaOmni Media
     - Teradata
   - GigaOmni Media




                         4
48th ACM Southeast Conference. ACMSE 2010.
Oxford, Mississippi. April 15-17, 2010.




                        Real-Time Analysis
                    of Streaming Sensor Data

                 Harshal Patni, Cory Henson, Michael Cooney,
                   AmitSheth, ThirunarayanKrishnaprasad
              Ohio Center of Excellence in Knowledge enabled Computing (Kno.e.sis)
                                Wright State University, Dayton, OH

                           Semantic Sensor Web @ Kno.e.sis
x
RT feature stream




                    Background Knowledge
                                           Rain Storm   Blizzard


                    ABSTRACTION



Huge amount of                             Features representing
Raw Sensor Data                              Real-World events




                            7
x
                                            Types of Abstractions
                                               Summarization across Thematic Dimension
Summarization over the Temporal Dimension




                                                                            8
x
Types of Abstractions
   Summarization across Thematic Dimension


                                                  Select


                                                    Join
                                                           Background
                                                           Knowledge
                                                Analyze




                                             Features representing
                                             Real-World Events
                                9
System Architecture
            x




                      10
An example problem?
                x

  “Find the sequence of weather events observed
    near Dayton James Cox Airport between
    Jan 13th and Jan 18th?”

    Thematic                  Spatial        Temporal

  Technologies required -
      1. Linked Sensor Data
      2. Feature Streams




                                        11
Sensor Discovery Application




                                       Weather Station ID

    Current Observations from
            MesoWest
                                                      Weather Station Coordinates


                                                    Weather Station Phenomena
    MesoWest – Project
    under Department of
    Meteorology, University
    of UTAH
                                              GeoNames – Geographic dataset


                  Sensors near Dayton James Cox Airport
                                       12
Linked Sensor Data




                             OGC
          MesoWest      Observation and                       Virtuoso RDF
                                               RDF Instance
         Service Data    Measurement                              store
                            (O&M)


                        O&M2RDFCONVERTER




                                          13
Summarizing Linked Sensor Data

         Extract Data for the sensor?              Find the sensor around
                                                   Dayton James Cox Airport?


                          procedure                       location
   Observation                                            location       Location KB
                                          Sensor KB
      KB                                                                  (Geonames)




                     procedure                            location
       720F                              Thermometer                     Dayton Airport




   • ~2 billion triples               • 20,000+ systems              • 230,000+ locations
   • MesoWest                         • MesoWest                     •Geonames
   •Static + Dynamic                  • ~Static                      • ~Static
x
Feature Composition




                      15
x
System Capability




                    16
x
System Feature Integration
 SELECT




            JOIN




                             17
Feature x
          Definition



• Rain Storm NOAA definition
 RainStorm = HighWindSpeed(above 35mph) AND
             Rain Precipitation AND
                                              SPARQL query for
             Temperature(greater than 32F)       RainStorm

                                                   Temperature


                                                      Rain
                                                   Precipitation

                                                    WindSpeed




                              18
FeaturexAnalysis



                        RDF Feature
                          Stream




                   19
Summarizing Feature Streams

                                      Feature Streams
                                            KB
                                                          Find sequence of events
                                                          near Dayton Airport?


                          procedure
   Observation                                            location       Location KB
                                          Sensor KB
      KB                                                                  (Geonames)




                     procedure                            location
       720F                              Thermometer                     Dayton Airport


   • ~2 billion triples               • 20,000+ systems              • 230,000+ locations
   • MesoWest                         • MesoWest                     •Geonames
   •Static + Dynamic                  • ~Static                      • ~Static
x
 Answering the query



“Find the sequence of weather events observed
  near Dayton James Cox Airport between
  Jan 13th and Jan 18th?”


  Linked Sensor Data          Feature Streams




                       21
x
     Demo




on-line video: http://www.youtube.com/watch?v=_ews4w_eCpg

                             22
Related Publications


  WORKSHOP PAPERS
  • Harshal Patni, Satya S. Sahoo, Cory Henson, Amit Sheth, Provenance Aware Linked Sensor
     Data, 2nd Workshop on Trust and Privacy on Social and Semantic Web,Co-Located with
     ESWC, Heraklion Greece, May 30th - June 3rd 2010
  • Harshal Patni, Cory Henson, Amit Sheth, Linked Sensor Data, In: Proceedings of 2010
     International Symposium on Collaborative Technologies and Systems (CTS 2010), Chicago, IL,
     May 17-21, 2010
  TECHNICAL REPORT
  • Harshal Patni, Cory Henson, Amit Sheth, and Pramod Ananthram. From Real Time Sensor
     Streams to Real Time Feature Streams, Kno.e.sis Center Technical Report, December 2009
  • Joshua Pschorr, Cory Henson, Harshal Patni, and Amit Sheth. Sensor Discovery on Linked
     Data, Kno.e.sis Center Technical Report, December 2009

  JOURNAL PAPER (In Progress)
  • Semantic Sensor Web: Design and Application towards weaving a meaningful sensor web



                                                23
Semantic Sensor Web




                      24
QUESTIONS


 Demos, Papers and more at: http://semantic-sensor-web.com
 Semantic Sensor Web @ Kno.e.sis




                                    25

Real-Time Analysis of Streaming Sensor Data

  • 2.
    x WEB DATA evolvedover time Real-Time Sensor, 2010’s Social, Multi-media data Dynamic User 2000’s Generated Content 1990’s Static Document and files 2
  • 3.
    x Properties of StreamingData Huge Volume Rapid Continuous Information Overload!! Heterogeneous 3
  • 4.
    x Some Statistics “A cross-country flight from New York to Los “Sensors Networksbeenproduce in the last the amount “More on a Boeing 737 plane 10-20 times massive will created Angeles data has generates a of generated -bythan in mediasummarize this data” Solution “Meaningfully in the next few years” three years social all the past 40,000 years” 240 terabytes of data” - GigaOmni Media - Teradata - GigaOmni Media 4
  • 5.
    48th ACM SoutheastConference. ACMSE 2010. Oxford, Mississippi. April 15-17, 2010. Real-Time Analysis of Streaming Sensor Data Harshal Patni, Cory Henson, Michael Cooney, AmitSheth, ThirunarayanKrishnaprasad Ohio Center of Excellence in Knowledge enabled Computing (Kno.e.sis) Wright State University, Dayton, OH Semantic Sensor Web @ Kno.e.sis
  • 6.
    x RT feature stream Background Knowledge Rain Storm Blizzard ABSTRACTION Huge amount of Features representing Raw Sensor Data Real-World events 7
  • 7.
    x Types of Abstractions Summarization across Thematic Dimension Summarization over the Temporal Dimension 8
  • 8.
    x Types of Abstractions Summarization across Thematic Dimension Select Join Background Knowledge Analyze Features representing Real-World Events 9
  • 9.
  • 10.
    An example problem? x “Find the sequence of weather events observed near Dayton James Cox Airport between Jan 13th and Jan 18th?” Thematic Spatial Temporal Technologies required - 1. Linked Sensor Data 2. Feature Streams 11
  • 11.
    Sensor Discovery Application Weather Station ID Current Observations from MesoWest Weather Station Coordinates Weather Station Phenomena MesoWest – Project under Department of Meteorology, University of UTAH GeoNames – Geographic dataset Sensors near Dayton James Cox Airport 12
  • 12.
    Linked Sensor Data OGC MesoWest Observation and Virtuoso RDF RDF Instance Service Data Measurement store (O&M) O&M2RDFCONVERTER 13
  • 13.
    Summarizing Linked SensorData Extract Data for the sensor? Find the sensor around Dayton James Cox Airport? procedure location Observation location Location KB Sensor KB KB (Geonames) procedure location 720F Thermometer Dayton Airport • ~2 billion triples • 20,000+ systems • 230,000+ locations • MesoWest • MesoWest •Geonames •Static + Dynamic • ~Static • ~Static
  • 14.
  • 15.
  • 16.
  • 17.
    Feature x Definition • Rain Storm NOAA definition RainStorm = HighWindSpeed(above 35mph) AND Rain Precipitation AND SPARQL query for Temperature(greater than 32F) RainStorm Temperature Rain Precipitation WindSpeed 18
  • 18.
    FeaturexAnalysis RDF Feature Stream 19
  • 19.
    Summarizing Feature Streams Feature Streams KB Find sequence of events near Dayton Airport? procedure Observation location Location KB Sensor KB KB (Geonames) procedure location 720F Thermometer Dayton Airport • ~2 billion triples • 20,000+ systems • 230,000+ locations • MesoWest • MesoWest •Geonames •Static + Dynamic • ~Static • ~Static
  • 20.
    x Answering thequery “Find the sequence of weather events observed near Dayton James Cox Airport between Jan 13th and Jan 18th?” Linked Sensor Data Feature Streams 21
  • 21.
    x Demo on-line video: http://www.youtube.com/watch?v=_ews4w_eCpg 22
  • 22.
    Related Publications WORKSHOP PAPERS • Harshal Patni, Satya S. Sahoo, Cory Henson, Amit Sheth, Provenance Aware Linked Sensor Data, 2nd Workshop on Trust and Privacy on Social and Semantic Web,Co-Located with ESWC, Heraklion Greece, May 30th - June 3rd 2010 • Harshal Patni, Cory Henson, Amit Sheth, Linked Sensor Data, In: Proceedings of 2010 International Symposium on Collaborative Technologies and Systems (CTS 2010), Chicago, IL, May 17-21, 2010 TECHNICAL REPORT • Harshal Patni, Cory Henson, Amit Sheth, and Pramod Ananthram. From Real Time Sensor Streams to Real Time Feature Streams, Kno.e.sis Center Technical Report, December 2009 • Joshua Pschorr, Cory Henson, Harshal Patni, and Amit Sheth. Sensor Discovery on Linked Data, Kno.e.sis Center Technical Report, December 2009 JOURNAL PAPER (In Progress) • Semantic Sensor Web: Design and Application towards weaving a meaningful sensor web 23
  • 23.
  • 24.
    QUESTIONS Demos, Papersand more at: http://semantic-sensor-web.com Semantic Sensor Web @ Kno.e.sis 25

Editor's Notes

  • #3 Web data evolved over time from Static documents -> dynamic user generated content (Web 2.0) -> streaming (real-time) dataSo what is real-time data and what are its properties
  • #4 Properties of streaming sensor data
  • #5 Social media is the dominant source of streaming data now, however in future sensors would …Data needs to be reduced
  • #6 Hence analyzing data in real-time becomes important
  • #7 Define a feature and introduce what do we mean by a feature
  • #8 So when we say Real-Time semantic analysis of streaming sensor data, we mean analyzing huge amount of lower level heterogeneous sensor data using background knowledge to create meaningful features (features that humans understand)
  • #9 Abstraction across the temporal dimension over a single stream vs abstraction across the thematic dimension over multiple, multimodal, heterogenous streams using background knowledge
  • #11 High level architecture
  • #12 To walk through the implementation, lets take a sample questionThe question might look trivial, but it contains 3 important parts
  • #13 Get all sensors using well known location names – Problem to be solveAssociate sensor descriptions to well know locations.
  • #14 Convert the data raw sensor data collected and convert to RDF using W3C’s sensor ontology
  • #19 Highlight the query with 3 boxes to show the temp, wind speed and precipitation streamHighlight the feature results too
  • #20 Talk about the observations and features storage
  • #22 Linked Data explodes
  • #23 Online demo – http://knoesis1.wright.edu/?q=rtfs