Transcript 7.FP7-SMART-FIA-Athens-180314-vFinal

Mobile Crowdsensing, Social and Big Data as Innovation Enablers for
Future Internet Cloud-based Architectures and Services
Integrating Social Sensors in IoT
applications for Smart Cities
John Soldatos, Associate Professor
Athens Information Technology
[email protected]
Athens - March 13, 2014
Internet-of-Things
Uniquely
Identifiable
Objects
Physical &
Virtual
Objects
Blend into
Business and
Social
Processes
Interoperable
Protocols
IoT
IoT Sensors (Data Sources) and
Actuators (Services)
IoT process data streams
(Physical & Virtual Sensors)
•
•
•
•
•
•
•
•
Cameras
Microphones
LIDAR
Radar
Energy Meter
Image Sensors
RFID / Barcode Reader
A/V Processing
Algorithms
• Manual Count / Human
Sensor
IoT invokes actuating
services
•
•
•
•
•
•
Motion Controller
Relays
Lights On/Off
Trigger Alarms
Start/Stop Device
Invoke Notification
Services
• Execute workflows
• Computers - Web
Services
«Social» Sensors
Types of «Social»
Sensors
• Sentiment
Analysis
• Topic-based
Community
Tracking
• Event Detection
• Opinion Mining
• More...
«Social» Sensors
Applications
•
•
•
•
•
Political Science
Market Research
Financial Services
Branding
Crisis
Management
• Law Enforcement
• More...
Citizens as «Social» Sensors
Citizens can act as sensors to connect with
governments and help the latter understand
their wishes and needs
Technologies
Typical Use
Cases
Use of Social
Media
GIS
Applications
Incident
Reporting
Tweet to
government
accounts @gov
Web and
Mobile Apps
Suggestions &
Comments
Access/post in
Facebook
pages etc.
IoT and Social Media
Social Media provide millions of insights on human activity and behaviour during
emergencies and security incidents
Examples: London Riots (Twitter), Egypt (Twitter/Facebook), but also «Sandy» Storm
(20M Tweets, 10 Instagram photos / sec)
IoT architectures deal
with the proliferating
«Social» Sensors
Relevant Technologies: Sentiment Analysis, Community Tracking, Rumour Spreading
Detection,...) - Used in several industries (marketing, branding, finance...)
IoT architectures and technologies support «Social» Sensors (as Virtual Sensor)
Twitter Sentiment Analysis On-line:
http://www.sentiment140.com/
Twitter Map During
«Sandy»
FP7 SMART Project Factsheet
Consortium:
Atos
Athens Information Technology
IBM Haifa Research Lab
Imperial College London
Consorzio S3LOG
TELESTO Technologies Ltd. (SME)
University of Glasgow (Research)
Prisa Digital
City-of-Santander
Timeframe: 01/11/2011-31/10/2014
Project Budget: 4.425.000 Euro
EC Contribution: 2.686.292 Euro
Web Site: http://www.smartfp7.eu/
FP7 SMART = Search Engine over Integrated
Social and Sensor Networks
MultiMedia
Social
• A/V Sensor Streams Processing
• Perceptive Components
• Intercepting Content from Social Networks
• Support for multiple social networks/media
Intelligent
• Fusion of Multiple Data Sour ces
• Reasoning over various Data Streams
Scalable &
Dynamic
• Novel Indexing & Retrieval Techniques
• Global Deployment (incl. BigData)
Open Source
• Based on Open Source Components (terrier.org)
• Publicly available based on an open source license
M.-D. Albakour, C. Macdonald, I. Ounis, A. Pnevmatikakis, J. Soldatos, «SMART: An Open Source Framework for
Searching the Physical World», Proceedings of the ACM SIGIR 2012 Workshop on Open Source Information.
SMART System Overview
SMART Edge Nodes
•Edge Node: SMART Point
of Presense
•Provides real-time
information from the city:
•Perceived from the
environment (sensors)
•Filtered from social networks
•Retrieved from the linked
data cloud
•Inferred by fusing the above
into higher-level events
Searching «Social» Feeds – Smart Reduce
System
Real-Time
Distributed
Architectures
Storm +
Terrier.org
SmartReduce
Prototype
Benchmarking
(Performance
& Scalability)
SMART Applications: Live News
Live News :
•“What is happening now?”
•“Which places are crowded?”,
•“What are the specific trends in
the city?”
•“Where are riots and fights
happening?”
•Answers = Multimedia streams
mixing sounds/images with
textual data stemming from
sensors and metadata steams
(including social networks)
•Deployed at City of Santander
SMART Applications: Security &
Surveillance
•Detect people and/or scenes that could be
considered as suspicious across certain times and
urban locations
•Data Streams from Cameras, Mics and Social
Networks
•Deployed at City of Santander
SMART Applications: Build Your Own
SC Scenario
•JSR 168 (Java Portlets)
compliant editor
•Sensor and Social Sensor
Driven Portlets and mashups
•Customized Portal
Developments for Smart City
Authorities
•Authoring tool for SMART
Cities Development
•Deployed at Santander,
Athens, Glasgow
SMART Open Source
http://opensoftware.smartfp7.eu/
SMART is open source
software released
Fully documented!
Users, Developers,
Contributors can access it at:
http://opensoftware.smartfp7.
eu/
Follow us: @smartfp7
Visit us at FIA Athens (2014)
Exhibition
Alternative Approach: Standardized
Semantic Modelling of «Social» Sensors
Semantic Interoperability
• Distributed and Heterogeneous Data
Sources
• Diverse Data Streams
• Common Semantics Needed
• Solution: Semantic Annoitation (W3C
Ontology)
Reasoning Algorithms
• Intelligent Selection & Filtering of Sensors
• Intelligent Selection & Filtering of Sensor
Data
• Use of Reasoners
• RDF/OWL Ontology (W3C SSN + Linked
Data)
Semantic Standards for sensors
provide a uniform way for
representing and reasoning over
heterogeneous data streams
«Social» Sensors in the Cloud
(OpenIoT project)
Performance
Capacity
Open Source Project
(https://github.com/
OpenIotOrg/openiot)
IoT in the
Cloud
Elasticity
Utility-Driven
Martin Serrano, Manfred Hauswirth, John Soldatos, Nikos Kefalakis, "Design Principles for Utility-Driven Services
and Cloud-Based Computing Modelling for the Internet of Things", International Journal of Web and Grid Services
(to appear), 2014.
Conclusions
Social Media Processing can provide millions of
insights on people and things attitude and behaviour
«Social» Sensors is a prominent Virtual Sensor in
the scope of IoT applications
IoT architectures make provisions for the integration
of Socia Media data
Semantic Interoperability of Social Media Streams is
a key for their dynamic discovery and use