The 4 V's of Big Data

My slides from the big data panel at GIScience 2012 in Columbus, OH.

Transcript

1. The 3 Vs Carsten Keßler | Institute for Geoinformatics | University of Münster, Germany http://carsten.io | @carstenkessler of Big Data 4
2. 1 | Volume LODUM Project LODUM Linked Open Data University of Münster
3. ‣ Integrating data of 70 different types (and counting): 1 | Volume LODUM Project LODUM Linked Open Data University of Münster
4. ‣ Integrating data of 70 different types (and counting): ‣ publications, persons, buildings, courses, cafeteria menus, urban history data, … 1 | Volume LODUM Project LODUM Linked Open Data University of Münster
5. ‣ Integrating data of 70 different types (and counting): ‣ publications, persons, buildings, courses, cafeteria menus, urban history data, … ‣ Challenge: reconciliation and de-duplication 1 | Volume LODUM Project LODUM Linked Open Data University of Münster
6. ‣ Integrating data of 70 different types (and counting): ‣ publications, persons, buildings, courses, cafeteria menus, urban history data, … ‣ Challenge: reconciliation and de-duplication > 800k resources, > 3 mio triples 1 | Volume LODUM Project LODUM Linked Open Data University of Münster
7. 2 | Velocity Semantic Sensor Networks / /
8. 2 | Velocity Semantic Sensor Networks ‣ Does Linked Sensor Data make sense at all? ‣ Do we want to provide all sensor data as Linked Sensor Data? / /
9. 2 | Velocity Semantic Sensor Networks ‣ Does Linked Sensor Data make sense at all? ‣ Do we want to provide all sensor data as Linked Sensor Data? ‣ Data streams ‣ Stream processing ‣ Automatically calibrate update frequency / /
10. 3 | Variety Humanitarian eXchange Language hxl.humanitarianresponse.info
11. 3 | Variety Humanitarian eXchange Language ‣ United Nations Office for the Coordination of Humanitarian Affairs (OCHA) hxl.humanitarianresponse.info
12. 3 | Variety Humanitarian eXchange Language ‣ United Nations Office for the Coordination of Humanitarian Affairs (OCHA) ‣ Integrating messy data from a wide range of humanitarian organizations in a number of formats hxl.humanitarianresponse.info
13. 3 | Variety Humanitarian eXchange Language ‣ United Nations Office for the Coordination of Humanitarian Affairs (OCHA) ‣ Integrating messy data from a wide range of humanitarian organizations in a number of formats ‣ Approach: ‣ semantic annotations ‣ space + time as organization principles hxl.humanitarianresponse.info
14. 4 | Veracity Trust and Reputation in VGI
15. 4 | Veracity Trust and Reputation in VGI
16. 4 | Veracity Trust and Reputation in VGI ‣ Parsing the history of single features ‣ Trying to compute credibility on a per-user basis
17. 4 | Veracity Trust and Reputation in VGI ‣ Parsing the history of single features ‣ Trying to compute credibility on a per-user basis ‣ Volume: ‣ Full OSM history dump: 500 GB ‣ Decompressing it takes ~36hours on a decent laptop
18. William Edwards Deming
19. In God we trust, all others bring data. William Edwards Deming