Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Semantics-empowered Smart City
Applications: Today and Tomorrow
Keynote at
The 6th Workshop on Semantics for Smarter Citie...
• Smart Cities as envisioned in the past and current state
• Smart City initiatives from governments
• Significance of dig...
• Smart Cities as envisioned in the past and current state
• Smart City initiatives from governments
• Significance of dig...
4
Source LAT Times, http://documents.latimes.com/la-2013/
Future Cities: A View from 1998
Thanks to Dr. Payam Barnaghi for...
5
Image courtesy: http://robertluisrabello.com/denial/traffic-in-la/#gallery[default]/0/
Source: wikipedia
Back to the Fut...
• Smart Cities as envisioned in the past and current state
• Smart City initiatives from governments
• Significance of dig...
7
https://www.whitehouse.gov/the-press-office/2015/09/14/fact-sheet-administration-announces-new-smart-cities-initiative-h...
8
Smart Cities: Significance and Impact
Image credit: https://commons.wikimedia.org/wiki/File:Narendra_Damodardas_Modi.jpg
• Smart Cities as envisioned in the past and current state
• Smart City initiatives from governments
• Significance of dig...
10
Smart Cities: A Historical Perspective
Economic development on trade routesCivilizations on river banks
Economic develo...
11
Smart City Applications: Proliferation of Digital Infrastructure
http://postscapes.com/internet-of-things-award/2014/sm...
12
Smart City Applications: Proliferation of Digital Infrastructure
http://postscapes.com/internet-of-things-award/2014/sm...
• Smart Cities as envisioned in the past and current state
• Smart City initiatives from governments
• Significance of dig...
14
Industrial SIE, ERIC
SME AI,
Higher
Education
UNIS, NUIG,
UASO, WSU
City BR, AA
Partners:
Duration: 36 months
CityPulse...
15
CityPulse
16
Data:
Data Processing Pipeline
17http://www.ict-citypulse.eu/scenarios/
101 Smart City Use-case Scenarios
18
Use-cases
19
http://www.ict-citypulse.eu/scenarios/
Use-case Scenarios
20
Scenario Ranking
• Smart Cities as envisioned in the past and current state
• Smart City initiatives from governments
• Significance of dig...
22
- Programmable devices
- Off-the-shelf gadgets/tools
Thanks to Dr. Payam Barnaghi for sharing the slide
Physical: Senso...
23
Thanks to Dr. Payam Barnaghi for sharing the slide
Cyber: Observations Pushed to the Cyber World
24
Motion sensor
Motion sensor
Motion sensor
ECG sensor
World Wide Web
Road block, A3
Road block, A3
Thanks to Dr. Payam B...
25
http://wiki.knoesis.org/index.php/PCS
Amit Sheth, Pramod Anantharam, Cory Henson, 'Physical-Cyber-Social Computing: An ...
• Smart Cities as envisioned in the past and current state
• Smart City initiatives from governments
• Significance of dig...
Image credit: http://www.ibm.com/smarterplanet/us/en/smarter_cities/overview/index.html
Public Safety Urban planning Gov. ...
Future of Smart Cities
29
Public Safety Urban planning Gov. & agency
admin.
Energy &
water
Environmental Transportation So...
Increased use of multimodal observations and knowledge for enhanced
explanation and prediction of city events to enable da...
31
Integrating Multimodal Observations: Transportation Domain Motivation
Link
Description
Scheduled
Event
Scheduled
Event
511.org
511.org
Schedule Information
511.org
32
Integrating Multimodal Ob...
• Why?
– Explain/Interpret average speed and link travel time data using event
schedule provided by city authorities and r...
• How?
– Extract events from textual tweets stream
– Build statistical models of normalcy, and thereby anomaly, from
numer...
• How?
– Extract events from textual tweets stream
– Build statistical models of normalcy, and thereby anomaly, from
numer...
• Are people talking about city traffic events on
twitter?
• Can we extract city traffic related events from
twitter?
• Ho...
37
Twitter as a Source of City Events
Public Safety
Urban planning
Gov. & agency
admin.
Energy & water
Environmental
Trans...
Some Challenges in Extracting Events from Tweets
• No well accepted definition of ‘events related to a city’
• Tweets are ...
39
Pramod Anantharam, Payam Barnaghi, Krishnaprasad Thirunarayan, and Amit Sheth. 2015. Extracting City Traffic Events fro...
• City Event Annotation
– Automated creation of training data
– Annotation task (our CRF model vs. baseline CRF model)
• C...
41
Distribution of Extracted Events Over Locations
• Evaluation Metric For Comparing Events with Ground Truth
– Complement...
42
Complementary Events
Complementary Events
Complementary Events
43
Corroborative Events
Corroborative Events
Corroborative Events
44
Timeliness
Timeliness
Evaluating Timeliness
• How?
– Extract events from textual tweets stream
– Build statistical models of normalcy, and thereby anomaly, from
numer...
Image credit: http://traffic.511.org/index
Multiple events
Varying influence
interact with each other
46
Challenge: Non-li...
• Causes of non-linearity in sensor data streams
– Temporal landmarks : peak hour vs off-peak traffic vs
weekend traffic
–...
• Disclaimer
"All models are wrong, but some are useful.” - George Box
• Normalcy Models
– Gaussian Mixture Model (GMM)
• ...
Image credit: http://tourontap.com/us-open-2012/courses-and-more-by-the-bay/
AT&T Park
49
Histogram of speed values
collected from June 1st 12:00 AM to June 2nd 12:00 AM
Histogram of travel time values
collected ...
Most of the drivers tend to
go 5 km/h over the posted speed limit
There are relatively few drivers who
go more than 10 km/...
52
Multiple Gaussian Distributions: A Better Fit for Speed Observations?
This distribution resembles a
Gaussian Mixture Mo...
Assume Normalcy to be uninterrupted traffic flow
July 2014 has no events so, we
hypothesize higher log-likelihood
score
Ju...
54
Hourly Traffic Dynamics Over a Day
55
Learning LDS Models
56
Tagging Anomalies with LDS Models
• How?
– Extract events from textual tweets stream
– Build statistical models of normalcy, and thereby anomaly, from
numer...
• If an anomaly is detected on a link L and during time
period [tst, tet], then the anomaly is explained by an event
if th...
• Data collected from San Francisco Bay Area between May 2014 to May
2015
– 511.org:
• 1,638 traffic incident reports
• 1....
• Smart Cities as envisioned in the past and current state
• Smart City initiatives from governments
• Significance of dig...
Innovative smart city solutions for situations with low/no instrumentation through
enhanced citizen participation
Coordina...
• May lead to second disaster to be managed:
– Under-supply of required demands
– Over-supply of not required resources
• ...
63
Image: http://www.gizmodo.com.au/2012/04/how-we-identify-single-
voices-in-a-crowd/
BIG QUESTION: Can these needles be ...
Really sparse Signal to Noise:
• 2M tweets during the first 48 hrs. of #Oklahoma-tornado-2013
- 1.3% as the precise resour...
Want to help animals in
#Oklahoma? @ASPCA
tells how you can help:
http://t.co/mt8l9PwzmO
x
RESPONSE TEAMS
(including human...
• Smart Cities as envisioned in the past and current state
• Smart City initiatives from governments
• Significance of dig...
Image Credit: http://www.webmd.com/news/breaking-news/future-of-health/#advances-in-vision-toc/advances-in-vision
http://w...
Proposed Ecosystem
- Kno.e.sis Center
- Manav Sadhna
- eMoksha
- Government
My son was on
Cloraquine for 2 days and
is not...
• Smart Cities as envisioned in the past and current state
• Smart City initiatives from governments
• Significance of dig...
• Utilizing multimodal and heterogeneous observations for
enhanced understanding and prediction of city events
• Create be...
• Dealing with massive heterogeneity in observations from a
city spanning physical, cyber, and social domains
• Dealing wi...
Thank You
http://knoesis.org/amit, http://personal.ee.surrey.ac.uk/Personal/P.Barnaghi/
@amit_p, @pbarnaghi
amit@knoesis.o...
Upcoming SlideShare
Loading in …5
×

Semantics-empowered Smart City applications: today and tomorrow

801 views

Published on

Citation:
Amit Sheth, "Semantics-empowered Smart City applications: today and tomorrow,” Keynote presented at the The 6th Workshop on Semantics for Smarter Cities (S4SC 2015), collocated with the 14th International Semantic Web Conference (ISWC2015), Bethlehem, PA, USA. Oct 11-12, 2015.
http://kat.ee.surrey.ac.uk/wssc/index.html

Abstract: There has been a massive growth in potentially relevant physical (sensor/IoT)- cyber (Web)- social data related to activities and operations of cities and citizens. As part of our participation in smart city projects, including the EU-funded CityPulse project, we have analyzed a large number of of use cases with inputs from city administrations and end users, and developed a few early applications. In this talk, I will present some exciting smart city applications possible today and venture to speculate on some future ones where Big Data technologies and semantic computing, including the use of domain knowledge, play a critical role.

Published in: Data & Analytics
  • Be the first to comment

Semantics-empowered Smart City applications: today and tomorrow

  1. 1. Semantics-empowered Smart City Applications: Today and Tomorrow Keynote at The 6th Workshop on Semantics for Smarter Cities (S4SC 2015), October 11-12, 2015 Prof. Amit Sheth LexisNexis Ohio Eminent Scholar; Executive Director, Kno.e.sis Wright State University Special Thanks: Pramod Anantharam http://www.ict-citypulse.eu/
  2. 2. • Smart Cities as envisioned in the past and current state • Smart City initiatives from governments • Significance of digital infrastructure for Smart Cities – Variety of smart city projects creating the digital infrastructure • CityPulse: Large-scale data analytics for smart cities • Key to Develop Future Robust Smart City Applications • Future of smart cities – P1: Increased use of multimodal observations and knowledge for enhanced explanation and prediction of city events to enable data driven policy making – P2: Innovative smart city solutions for situations with low/no instrumentation through seamless citizen participation • ComSmart Cities: opportunities and challenges – P1: Integrating Multimodal Observations: Transportation Domain • Combining textual traffic events with traffic sensor data – P2: Smart city solutions for situations with low/no instrumentation • Coordination during Disasters • Smart Health for Situations with Low/No Instrumentation • Smart Cities: opportunities and challenges Outline
  3. 3. • Smart Cities as envisioned in the past and current state • Smart City initiatives from governments • Significance of digital infrastructure for Smart Cities – Variety of smart city projects creating the digital infrastructure • CityPulse: Large-scale data analytics for smart cities • Key to Develop Future Robust Smart City Applications • Future of smart cities – P1: Increased use of multimodal observations and knowledge for enhanced explanation and prediction of city events to enable data driven policy making – P2: Innovative smart city solutions for situations with low/no instrumentation through seamless citizen participation • ComSmart Cities: opportunities and challenges – P1: Integrating Multimodal Observations: Transportation Domain • Combining textual traffic events with traffic sensor data – P2: Smart city solutions for situations with low/no instrumentation • Coordination during Disasters • Smart Health for Situations with Low/No Instrumentation • Smart Cities: opportunities and challenges Outline
  4. 4. 4 Source LAT Times, http://documents.latimes.com/la-2013/ Future Cities: A View from 1998 Thanks to Dr. Payam Barnaghi for sharing the slide
  5. 5. 5 Image courtesy: http://robertluisrabello.com/denial/traffic-in-la/#gallery[default]/0/ Source: wikipedia Back to the Future: 2013
  6. 6. • Smart Cities as envisioned in the past and current state • Smart City initiatives from governments • Significance of digital infrastructure for Smart Cities – Variety of smart city projects creating the digital infrastructure • CityPulse: Large-scale data analytics for smart cities • Key to Develop Future Robust Smart City Applications • Future of smart cities – P1: Increased use of multimodal observations and knowledge for enhanced explanation and prediction of city events to enable data driven policy making – P2: Innovative smart city solutions for situations with low/no instrumentation through seamless citizen participation • ComSmart Cities: opportunities and challenges – P1: Integrating Multimodal Observations: Transportation Domain • Combining textual traffic events with traffic sensor data – P2: Smart city solutions for situations with low/no instrumentation • Coordination during Disasters • Smart Health for Situations with Low/No Instrumentation • Smart Cities: opportunities and challenges Outline
  7. 7. 7 https://www.whitehouse.gov/the-press-office/2015/09/14/fact-sheet-administration-announces-new-smart-cities-initiative-help Image credit: https://www.whitehouse.gov/administration/president-obama Smart Cities: Significance and Impact
  8. 8. 8 Smart Cities: Significance and Impact Image credit: https://commons.wikimedia.org/wiki/File:Narendra_Damodardas_Modi.jpg
  9. 9. • Smart Cities as envisioned in the past and current state • Smart City initiatives from governments • Significance of digital infrastructure for Smart Cities – Variety of smart city projects creating the digital infrastructure • CityPulse: Large-scale data analytics for smart cities • Key to Develop Future Robust Smart City Applications • Future of smart cities – P1: Increased use of multimodal observations and knowledge for enhanced explanation and prediction of city events to enable data driven policy making – P2: Innovative smart city solutions for situations with low/no instrumentation through seamless citizen participation • ComSmart Cities: opportunities and challenges – P1: Integrating Multimodal Observations: Transportation Domain • Combining textual traffic events with traffic sensor data – P2: Smart city solutions for situations with low/no instrumentation • Coordination during Disasters • Smart Health for Situations with Low/No Instrumentation • Smart Cities: opportunities and challenges Outline
  10. 10. 10 Smart Cities: A Historical Perspective Economic development on trade routesCivilizations on river banks Economic development now increasingly rely on digital infrastructure 10 Image credit: http://www.rcet.org/twd/students/socialstudies/ss_extensions_1intro.html Image credit: http://www.shutterstock.com/pic-157118819/stock-vector-conceptual-tag-cloud-containing-words-related-to-smart-city-digital-city-infrastructure-ict.html
  11. 11. 11 Smart City Applications: Proliferation of Digital Infrastructure http://postscapes.com/internet-of-things-award/2014/smart-city-application.html
  12. 12. 12 Smart City Applications: Proliferation of Digital Infrastructure http://postscapes.com/internet-of-things-award/2014/smart-city-application.html
  13. 13. • Smart Cities as envisioned in the past and current state • Smart City initiatives from governments • Significance of digital infrastructure for Smart Cities – Variety of smart city projects creating the digital infrastructure • CityPulse: Large-scale data analytics for smart cities • Key to Develop Future Robust Smart City Applications • Future of smart cities – P1: Increased use of multimodal observations and knowledge for enhanced explanation and prediction of city events to enable data driven policy making – P2: Innovative smart city solutions for situations with low/no instrumentation through seamless citizen participation • ComSmart Cities: opportunities and challenges – P1: Integrating Multimodal Observations: Transportation Domain • Combining textual traffic events with traffic sensor data – P2: Smart city solutions for situations with low/no instrumentation • Coordination during Disasters • Smart Health for Situations with Low/No Instrumentation • Smart Cities: opportunities and challenges Outline
  14. 14. 14 Industrial SIE, ERIC SME AI, Higher Education UNIS, NUIG, UASO, WSU City BR, AA Partners: Duration: 36 months CityPulse: Large-scale Data Analytics for Smart Cities
  15. 15. 15 CityPulse
  16. 16. 16 Data: Data Processing Pipeline
  17. 17. 17http://www.ict-citypulse.eu/scenarios/ 101 Smart City Use-case Scenarios
  18. 18. 18 Use-cases
  19. 19. 19 http://www.ict-citypulse.eu/scenarios/ Use-case Scenarios
  20. 20. 20 Scenario Ranking
  21. 21. • Smart Cities as envisioned in the past and current state • Smart City initiatives from governments • Significance of digital infrastructure for Smart Cities – Variety of smart city projects creating the digital infrastructure • CityPulse: Large-scale data analytics for smart cities • Key to Develop Future Robust Smart City Applications • Future of smart cities – P1: Increased use of multimodal observations and knowledge for enhanced explanation and prediction of city events to enable data driven policy making – P2: Innovative smart city solutions for situations with low/no instrumentation through seamless citizen participation • ComSmart Cities: opportunities and challenges – P1: Integrating Multimodal Observations: Transportation Domain • Combining textual traffic events with traffic sensor data – P2: Smart city solutions for situations with low/no instrumentation • Coordination during Disasters • Smart Health for Situations with Low/No Instrumentation • Smart Cities: opportunities and challenges Outline
  22. 22. 22 - Programmable devices - Off-the-shelf gadgets/tools Thanks to Dr. Payam Barnaghi for sharing the slide Physical: Sensors Monitoring the Physical World
  23. 23. 23 Thanks to Dr. Payam Barnaghi for sharing the slide Cyber: Observations Pushed to the Cyber World
  24. 24. 24 Motion sensor Motion sensor Motion sensor ECG sensor World Wide Web Road block, A3 Road block, A3 Thanks to Dr. Payam Barnaghi for sharing the slide Social: People Interacting with the Physical World
  25. 25. 25 http://wiki.knoesis.org/index.php/PCS Amit Sheth, Pramod Anantharam, Cory Henson, 'Physical-Cyber-Social Computing: An Early 21st Century Approach,' IEEE Intelligent Systems, vol. 28, no. 1, pp. 78-82, Jan.-Feb., 2013. http://doi.ieeecomputersociety.org/10.1109/MIS.2013.20 Physical Cyber Social* Developers need to Consider observations from Physical-Cyber-Social systems in building future Smart City applications *http://www.ichangemycity.com/ Key to Develop Future Robust Smart City Applications
  26. 26. • Smart Cities as envisioned in the past and current state • Smart City initiatives from governments • Significance of digital infrastructure for Smart Cities – Variety of smart city projects creating the digital infrastructure • CityPulse: Large-scale data analytics for smart cities • Key to Develop Future Robust Smart City Applications • Future of smart cities – P1: Increased use of multimodal observations and knowledge for enhanced explanation and prediction of city events to enable data driven policy making – P2: Innovative smart city solutions for situations with low/no instrumentation through seamless citizen participation • ComSmart Cities: opportunities and challenges – P1: Integrating Multimodal Observations: Transportation Domain • Combining textual traffic events with traffic sensor data – P2: Smart city solutions for situations with low/no instrumentation • Coordination during Disasters • Smart Health for Situations with Low/No Instrumentation • Smart Cities: opportunities and challenges Outline
  27. 27. Image credit: http://www.ibm.com/smarterplanet/us/en/smarter_cities/overview/index.html Public Safety Urban planning Gov. & agency admin. Energy & water Environmental Transportation Social Programs Healthcare Education Some Smart City Components 28
  28. 28. Future of Smart Cities 29 Public Safety Urban planning Gov. & agency admin. Energy & water Environmental Transportation Social Programs Healthcare Education • Increased use of multimodal observations and knowledge for enhanced explanation and prediction of city events to enable data driven policy making • Innovative smart city solutions for situations with low/no instrumentation through seamless citizen participation
  29. 29. Increased use of multimodal observations and knowledge for enhanced explanation and prediction of city events to enable data driven policy making Multimodal observations and knowledge Enhanced explanation and prediction Data driven policy decisions Public Safety Urban planning Gov. & agency admin. Energy & water Environmental Transportation Social Programs Healthcare Education https://www.oracle.com/applications/enterprise-resource-planning/roles/chief-financial-officer.html
  30. 30. 31 Integrating Multimodal Observations: Transportation Domain Motivation
  31. 31. Link Description Scheduled Event Scheduled Event 511.org 511.org Schedule Information 511.org 32 Integrating Multimodal Observations: Transportation Domain Motivation Slow moving traffic
  32. 32. • Why? – Explain/Interpret average speed and link travel time data using event schedule provided by city authorities and real-time traffic events shared on Twitter – Past work: Predict congestion using single modality such as sensor data • What? – Combine • 511.org data about Bay Area Road Network Traffic – E.g., Average speed and link travel time data stream – E.g., (Happened or planned) event reports • Tweets that report events including ad hoc ones 33 Integrating Multimodal Observations: Transportation Domain Thanks to Dr. Krishnaprasad Thirunarayan for sharing the slide
  33. 33. • How? – Extract events from textual tweets stream – Build statistical models of normalcy, and thereby anomaly, from numerical sensor data streams – Correlate multimodal streams, using spatio-temporal information, to annotate “anomalies” in sensor data time series with textual events 34 Integrating Multimodal Observations: Transportation Domain Thanks to Dr. Krishnaprasad Thirunarayan for sharing the slide
  34. 34. • How? – Extract events from textual tweets stream – Build statistical models of normalcy, and thereby anomaly, from numerical sensor data streams – Correlate multimodal streams, using spatio-temporal information, to annotate “anomalies” in sensor data time series with textual events 35 Integrating Multimodal Observations: Transportation Domain
  35. 35. • Are people talking about city traffic events on twitter? • Can we extract city traffic related events from twitter? • How can we leverage event and location knowledge bases for event extraction? • How well can we extract city events? Research Questions 36
  36. 36. 37 Twitter as a Source of City Events Public Safety Urban planning Gov. & agency admin. Energy & water Environmental TransportationSocial Programs Healthcare Education
  37. 37. Some Challenges in Extracting Events from Tweets • No well accepted definition of ‘events related to a city’ • Tweets are short (140 characters) and its informal nature make it hard to analyze – Entity, location, time, and type of the event • Multiple reports of the same event and sparse report of some events (biased sample) – Numbers don’t necessarily indicate intensity • Validation of the solution is hard due to the open domain nature of the problem 38
  38. 38. 39 Pramod Anantharam, Payam Barnaghi, Krishnaprasad Thirunarayan, and Amit Sheth. 2015. Extracting City Traffic Events from Social Streams. ACM Trans. Intell. Syst. Technol. 6, 4, Article 43 (July 2015), 27 pages. DOI=10.1145/2717317 http://doi.acm.org/10.1145/2717317 Extracting City Events from Textual Data
  39. 39. • City Event Annotation – Automated creation of training data – Annotation task (our CRF model vs. baseline CRF model) • City Event Extraction – Use aggregation algorithm for event extraction – Extracted events AND ground truth • Dataset (Aug – Nov 2013) ~ 8 GB of data on disk – Over 8 million tweets – Over 162 million sensor data points – 311 active events and 170 scheduled events First Evaluation 40
  40. 40. 41 Distribution of Extracted Events Over Locations • Evaluation Metric For Comparing Events with Ground Truth – Complementary Events • Additional information e.g., slow traffic from sensor data and accident from textual data – Corroborative Events • Additional confidence e.g., accident event supporting a accident report from ground truth – Timeliness • Early detection e.g., knowing poor visibility before its formal report
  41. 41. 42 Complementary Events Complementary Events Complementary Events
  42. 42. 43 Corroborative Events Corroborative Events Corroborative Events
  43. 43. 44 Timeliness Timeliness Evaluating Timeliness
  44. 44. • How? – Extract events from textual tweets stream – Build statistical models of normalcy, and thereby anomaly, from numerical sensor data streams – Correlate multimodal streams, using spatio-temporal information, to annotate “anomalies” in sensor data time series with textual events 45 Traffic Domain Use-case (Open Data)
  45. 45. Image credit: http://traffic.511.org/index Multiple events Varying influence interact with each other 46 Challenge: Non-linearity in Traffic Dynamics
  46. 46. • Causes of non-linearity in sensor data streams – Temporal landmarks : peak hour vs off-peak traffic vs weekend traffic – Effect of location – Scheduled events such as road construction, baseball game, or music concert – Unexpected events such as accidents or heavy rains – Random variations (viz., stochasticity) 47 Traffic Dependencies Thanks to Dr. Krishnaprasad Thirunarayan for sharing the slide
  47. 47. • Disclaimer "All models are wrong, but some are useful.” - George Box • Normalcy Models – Gaussian Mixture Model (GMM) • Captures multiple co-existing events and its impact on traffic • Doesn’t capture temporal dependencies – Auto Regressive (AR) Models • Captures temporal dependencies in traffic dynamics • Doesn’t capture hidden aspects of the domain (e.g., volume of traffic) – Linear Dynamical System (LDS) • Captures temporal dependencies and hidden aspects of a domain • Anomaly Model – Cf. Box and Whisker plots 48 Abstracting Traffic Behavior: Traffic Data Model Thanks to Dr. Krishnaprasad Thirunarayan for sharing the slide
  48. 48. Image credit: http://tourontap.com/us-open-2012/courses-and-more-by-the-bay/ AT&T Park 49
  49. 49. Histogram of speed values collected from June 1st 12:00 AM to June 2nd 12:00 AM Histogram of travel time values collected from June 1st 12:00 AM to June 2nd 12:00 AM 50 Traffic Data: First Peek
  50. 50. Most of the drivers tend to go 5 km/h over the posted speed limit There are relatively few drivers who go more than 10 km/h over the posted speed limit There are situations in a day where the drivers are going (forced) below the speed limit e.g., rush hour traffic Do these histograms resemble any probability distribution? 51 Traffic Data: Possible Explanation
  51. 51. 52 Multiple Gaussian Distributions: A Better Fit for Speed Observations? This distribution resembles a Gaussian Mixture Model (GMM)
  52. 52. Assume Normalcy to be uninterrupted traffic flow July 2014 has no events so, we hypothesize higher log-likelihood score June 2014 has many events so, we hypothesize lower log-likelihood score -115655.8 (Closer to Normalcy) -125974.3 53 Golden Gate Fields: Comparing Months with Varying Event Occurrences
  53. 53. 54 Hourly Traffic Dynamics Over a Day
  54. 54. 55 Learning LDS Models
  55. 55. 56 Tagging Anomalies with LDS Models
  56. 56. • How? – Extract events from textual tweets stream – Build statistical models of normalcy, and thereby anomaly, from numerical sensor data streams – Correlate multimodal streams, using spatio-temporal information, to annotate “anomalies” in sensor data time series with textual events 57 Traffic Domain Use-case (Open Data)
  57. 57. • If an anomaly is detected on a link L and during time period [tst, tet], then the anomaly is explained by an event if the event occurred in the vicinity within 0.5km radius and during [tst-1, tet+1]. • CAVEAT: An anomaly may not be explained because of missing data. 58 Spatio-temporal Co-occurrence Criteria Thanks to Dr. Krishnaprasad Thirunarayan for sharing the slide
  58. 58. • Data collected from San Francisco Bay Area between May 2014 to May 2015 – 511.org: • 1,638 traffic incident reports • 1.4 billion speed and travel time observations – Twitter Data: 39,208 traffic related incidents extracted from over 20 million tweets1 • Naïve implementation for learning normalcy models for 2,534 links resulted in 40 minutes per link (~ 2 months of processing time for our data) – 2.66 GHz, Intel Core 2 Duo with 8 GB main memory • Scalable implementation by exploiting the nature of the problem resulted in learning normalcy models within 24 hours – The Apache Spark cluster used in our evaluation has 864 cores and 17TB main memory. 59 1Anantharam, P. 2014. Extracting city traffic events from social streams. https://osf.io/b4q2t/wiki/home/ Experimental Data Statistics and Infrastructure
  59. 59. • Smart Cities as envisioned in the past and current state • Smart City initiatives from governments • Significance of digital infrastructure for Smart Cities – Variety of smart city projects creating the digital infrastructure • CityPulse: Large-scale data analytics for smart cities • Key to Develop Future Robust Smart City Applications • Future of smart cities – P1: Increased use of multimodal observations and knowledge for enhanced explanation and prediction of city events to enable data driven policy making – P2: Innovative smart city solutions for situations with low/no instrumentation through seamless citizen participation • ComSmart Cities: opportunities and challenges – P1: Integrating Multimodal Observations: Transportation Domain • Combining textual traffic events with traffic sensor data – P2: Smart city solutions for situations with low/no instrumentation • Coordination during Disasters • Smart Health for Situations with Low/No Instrumentation • Smart Cities: opportunities and challenges Outline
  60. 60. Innovative smart city solutions for situations with low/no instrumentation through enhanced citizen participation Coordination during Disasters
  61. 61. • May lead to second disaster to be managed: – Under-supply of required demands – Over-supply of not required resources • Hurricane Sandy example, “Thanks, but no thanks”, NPR, Jan 12 2013 Story link: http://www.npr.org/2013/01/09/168946170/thanks- but-no-thanks-when-post-disaster-donations- overwhelm Uncoordinated Engagement
  62. 62. 63 Image: http://www.gizmodo.com.au/2012/04/how-we-identify-single- voices-in-a-crowd/ BIG QUESTION: Can these needles be identified in the haystack of massive datasets? Me and @CeceVancePR are coordinating a clothing/food drive for families affected by Hurricane Sandy. If you would like to donate, DM us Does anyone know how to donate clothes to hurricane #Sandy victims? [REQUEST/DEMAND] [OFFER/SUPPLY] Coordination teams want to hear! [BIG] Ad-hoc Community with Varying but [FEW] Important Intents
  63. 63. Really sparse Signal to Noise: • 2M tweets during the first 48 hrs. of #Oklahoma-tornado-2013 - 1.3% as the precise resource donation requests to help - 0.02% as the precise resource donation offers to help 64 • Anyone know how to get involved to help the tornado victims in Oklahoma??#tornado #oklahomacity (OFFER) • I want to donate to the Oklahoma cause shoes clothes even food if I can (OFFER) Disaster Response Coordination: Finding Actionable Nuggets for Responders to act • Text REDCROSS to 909-99 to donate to those impacted by the Moore tornado! http://t.co/oQMljkicPs (REQUEST) • Please donate to Oklahoma disaster relief efforts.: http://t.co/crRvLAaHtk (REQUEST) For responders, most important information is the scarcity and availability of resources Blog by our colleague Patrick Meier on this analysis: http://irevolution.net/2013/05/29/analyzing-tweets-tornado/
  64. 64. Want to help animals in #Oklahoma? @ASPCA tells how you can help: http://t.co/mt8l9PwzmO x RESPONSE TEAMS (including humanitarian org. and ‘pseudo’ responders) VICTIM SITE Where do I go to help out for volunteer work around Moore? Anyone know? Anyone know where to donate to help the animals from the Oklahoma disaster? #oklaho ma #dogs Matchable Matchable If you would like to volunteer today, help is desperately needed in Shawnee. Call 273-5331 for more info 65 CITIZEN SENSORS DEMAND SUPPLY Match-making: Assisting Coordination Image: http://offthewallsocial.com/tag/social-media/
  65. 65. • Smart Cities as envisioned in the past and current state • Smart City initiatives from governments • Significance of digital infrastructure for Smart Cities – Variety of smart city projects creating the digital infrastructure • CityPulse: Large-scale data analytics for smart cities • Key to Develop Future Robust Smart City Applications • Future of smart cities – P1: Increased use of multimodal observations and knowledge for enhanced explanation and prediction of city events to enable data driven policy making – P2: Innovative smart city solutions for situations with low/no instrumentation through seamless citizen participation • ComSmart Cities: opportunities and challenges – P1: Integrating Multimodal Observations: Transportation Domain • Combining textual traffic events with traffic sensor data – P2: Smart city solutions for situations with low/no instrumentation • Coordination during Disasters • Smart Health for Situations with Low/No Instrumentation • Smart Cities: opportunities and challenges Outline
  66. 66. Image Credit: http://www.webmd.com/news/breaking-news/future-of-health/#advances-in-vision-toc/advances-in-vision http://www.georgeinstitute.org/philanthropic-opportunities/smart-health-india http://transmissionsmedia.com/wp-content/uploads/2013/02/NGO.jpg https://developer.ibm.com/apimanagement/2014/11/20/government-apis-less/ Better understanding for policy decisions Technology enabled enhanced citizen participation Smart Health for Situations with Low/No Instrumentation Innovative smart city solutions for situations with low/no instrumentation through enhanced citizen participation Seamless interactions b/w citizens, NGOs, and Governments • Localize symptoms and disease prevalence • Localize environmental conditions e.g., presence of mosquitos, stagnant water • Understand epidemics and its propagation • Measures to curb epidemics by timely supply of medication and advice Heterogeneous data acquisition from healthcare workers and low-cost sensors (e.g., SMS, and images).
  67. 67. Proposed Ecosystem - Kno.e.sis Center - Manav Sadhna - eMoksha - Government My son was on Cloraquine for 2 days and is not showing any improvements on malaria symptoms. Our resources will not last long if the malaria cases increase in a few days. We are in need of medications and volunteers. Small water pools around the neighborhood are creating mosquito problems. User 1 User 2 User 3 Chloraquine is not a suggested solution for malaria in India. Please see a provider ASAP. The closest healthcare facility is on street X. Received several comments from this area regarding malaria symptoms. Please send your volunteer to check. Query from User 1: classified as an “active care” type query. Response needs to be sent to the user. Further analysis showed similar queries from same region. Query from User 2 and 3: classified as a “preventive care” query. The message needs to be sent to an NGO. Water pool is breeding site for Anopheles mosquitos, so preventive measures need to be taken. SMS, e-mails, tweets, Web People from various locations Ontology/Knowledg e Base
  68. 68. • Smart Cities as envisioned in the past and current state • Smart City initiatives from governments • Significance of digital infrastructure for Smart Cities – Variety of smart city projects creating the digital infrastructure • CityPulse: Large-scale data analytics for smart cities • Key to Develop Future Robust Smart City Applications • Future of smart cities – P1: Increased use of multimodal observations and knowledge for enhanced explanation and prediction of city events to enable data driven policy making – P2: Innovative smart city solutions for situations with low/no instrumentation through seamless citizen participation • ComSmart Cities: opportunities and challenges – P1: Integrating Multimodal Observations: Transportation Domain – Combining textual traffic events with traffic sensor data – P2: Smart city solutions for situations with low/no instrumentation • Coordination during Disasters • Smart Health for Situations with Low/No Instrumentation • Smart Cities: opportunities and challenges Outline
  69. 69. • Utilizing multimodal and heterogeneous observations for enhanced understanding and prediction of city events • Create better governance of our cities and better public services through data driven policy making • Empower citizens for active participation in shaping the development of a city • Provide more business opportunities for companies (and SMEs) and private sector services • Improve energy efficiency, create greener environments… • Create better healthcare, elderly-care… Thanks to Dr. Payam Barnaghi for sharing the slide 70 Smart Cities: Opportunities
  70. 70. • Dealing with massive heterogeneity in observations from a city spanning physical, cyber, and social domains • Dealing with missing, sparse, and noisy observations from machine sensors and people • Seamless integration of citizens in shaping city policies (reliability and quality of citizen reporting of city events) • Reliability and dependability of the massive infrastructure of connected devices, services, and people • Transparency and data management issues (privacy, security, trust, …) 71 Smart Cities: Challenges
  71. 71. Thank You http://knoesis.org/amit, http://personal.ee.surrey.ac.uk/Personal/P.Barnaghi/ @amit_p, @pbarnaghi amit@knoesis.org, p.barnaghi@surrey.ac.uk Acknowledgement: CityPulse Consortium http://www.ict-citypulse.eu Annual Report: http://www.ict-citypulse.eu/page/sites/default/files/citypulse_annual_report.pdf Kno.e.sis – Ohio Center of Excellence in Knowledge-enabled Computing Wright State University, Dayton, Ohio, USA

×