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.

Das QROWD-Projekt - Because Big Data Integration is Humanly Possible

17 views

Published on

LSWT2019 Talk by Simon Bin, AKSW/InfAI e.V.; Patrick Westphal, AKSW/InfAI e.V.

Published in: Technology
  • Be the first to comment

  • Be the first to like this

Das QROWD-Projekt - Because Big Data Integration is Humanly Possible

  1. 1. The QROWD Project H2020 Project, December 2016 - December 2019 LSWT 2019 Simon Bin, Patrick Westphal, Claus Stadler, Gordian Dziwis (InfAI) This project has received funding from the European Union’s Horizon 2020 research and innovation programme
  2. 2. Motivation 2  Information services {[]} CSV <?xml …> … “Data at rest” “Data in motion” Data Integration ● Efficient data-driven transportation and mobility ● Reduce CO2 emissions ● Assess the quality of infrastructure ● Enhance the quality of life of citizens Data Analytics Inform Optimize 2
  3. 3. The Consortium 3  - Business technology company - Data generation and acquisition - In-car location and navigation company - Data and service provider - Software architecture and Semantic Web company - Architecture design and integration of QROWD platform - Business and marketing consulting company - Dissemination, communication and exploitation - University of Southampton - Human computation, crowdsourcing and citizen science - University of Trento - Mobile app for citizen sensing/crowd feedback - Central storage infrastructure - Research institute @Leipzig University - Data transformation and integration - Analytics - Municipality of Trento - Data provider - End user role 3
  4. 4. The Model Region 4  City of Trento ● Capital of Trentino region (Italy) ● People: 120.000 (2017) ● Third largest city in the Alps; second largest in the Tyrol ● Smartest mid-sized city in Italy (iCity Rate, 2017) 4
  5. 5. The Overall Data Flow 5 QROWD citizen feedback component {[]} CSV <?xml …> … … Links Represent- atives QROWD Platform … ??? Questions/answers DBpedia, LinkedgeoData, ... LOD Cloud QROWD data conversion component (SparqlIntegrate, ...) QROWD analytics components (DL-Learner, ...) QROWD linking and fusion components (LIMES, …)
  6. 6. Data Acquisition 6  ● SparqlIntegrate 6 process.sparqlgenerate-workloads.script workloads.sparql emit.sparql distributions generates references adds iterates workloads selects output triples yields references
  7. 7. 7  ● 7 Linking and Fusion Use Case ‘Bike Rack’
  8. 8. Linking and Fusion Use Case ‘Bike Rack’ 8  ● Linking on geodata 8
  9. 9. Analytics Use Case ‘Modal Split’ 9  9 ● Task: Given commute information captured on citizens’ smartphones + (geospatial) background knowledge, detect the citizens’ transportation mode ● Detected modes can be aggregated → ‘Modal split’ ● Distinctive patterns based on implicit spatial relations (near, starts near, ends near, runs along, passes, is within, crosses, …) ○ Commute ran along a line feature of type Railway→ transportation mode is probably ‘train’ ○ Commute started and ended near POI of type BusStop → transportation mode is probably ‘bus’ ○ …
  10. 10. Analytics Use Case ‘Modal Split’ 10 QROWD citizen feedback component {[]} CSV <?xml …> … QROWD data conversion component (SparqlIntegrate, ...) … QROWD analytics components (DL-Learner, ...) - OWL Reasoning (via off-the-shelf OWL reasoner + optimizations) - Spatial learning extension - Temporal learning extension QROWD linking and fusion components (LIMES, …) Links Represent- atives QROWD mobile app for citizen sensing and feedback (i-Log) QROWD Platform …GPS data ??? … Questions/answers DBpedia, LinkedgeoData, ... LOD Cloud Trento-specific opendatasets
  11. 11. https://www.qrowd-project.eu @QrowdProject https://github.com/QROWD Thank you!
  12. 12. Analytics Use Case ‘Modal Split’ 12 spatialReasoner.getSpatialIndividualsOnWhichRunsAlong(bikeRide) > [ <http://linkedgeodata.org/triplify/way257254437>, <http://linkedgeodata.org/triplify/way413052887>, <http://linkedgeodata.org/triplify/way493758696>, <http://linkedgeodata.org/triplify/way257254437> ]

×