Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Do you really want to delete this prezi?
Neither you, nor the coeditors you shared it with will be able to recover it again.
Make your likes visible on Facebook?
Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.
Crowd Sourced Data - It can't be that accurate can it?
Leigh Hunton 6 July 2011
Transcript of Crowd Sourced Data - It can't be that accurate can it?
Splice Group Inaugural 2011 ALGIM GIS Symposium What do we mean by Accurate? Spatial accuracy
Timeliness What do we mean by... Source: Haklay 2010 (ii), (attributed to P A J Van Oort, 2006) Sources Haklay 2010 - How Many Volunteers Does it Take to Map an Area Well? The Validaty of Linus' Law to Volunteered Geographic Information
Mordechai Haklay, Sofia Badiouka, Vyron Antonion and Aamer Ather
The Cartograophic Journal Vol. 47 No. 4 pages 315-322 November 2010
Haklay 2010 (ii)- How good is volunteered geographical information?
A comparitive study of OpenStreetMap and Ordnance Survey datasets
Environmental and Planning B: Planning and Design 2010, Vol. 37, pages 682-703 140,000 tweets processed |< 12:51 #eqnz hashtag |< +2h30m Collaboration moved to IRC and Skype... Call for help echoed over next few hours... |< +3h30min 50 technical volunteers involved (Not all tweets impart crucial information) QA Tools
Detection of changes
Visualisation of changes How do we measure Quality? Many things...
Become part of a larger community
Disatisfaction with traditional model
Having their voice heard
Significant Events What motivates the "Crowd"? http://www.flickr.com/photos/kwl/4116389731 RRRROAR! I have some very constructive advice I would like to share with you on how to improve your systems and processes. |< +90m Collaboration taking form |< +1h40m Initially shared notepads 12:00 | 13:00 | 14:00 | 15:00 | 16:00 | 17:00 | 18:00 | 19:00 | 20:00 | 21:00 | 22:00 ...and CrisisCampNZ Attribute Accuracy Completeness Timeliness Spatial Accuracy |< 20 m >| 40 GPS traces Data courtesy: Napier City Council |< 57 m >| Napier CBD >|5m|< >|5m|< >|10 m|< Aerial misalignment Realign aerials against GPS traces Spatial Accuracy Data courtesy: Napier City Council. Aerial Photography: Bing Data courtesy: Napier City Council accuracy +/- 0.3m
Captured and processed by New Zealand Aerial Mapping Data courtesy: Napier City Council OSM - incomplete roundabout
Bing and OSM positional accuracy is similar
Need a better reference set to compare against Links Data courtesy: Napier City Council |< +9hr Spread around globe |< +5h Being reported in National Press source: http://commons.wikimedia.org/wiki/File:Phone-sony-ericsson-j100i.svg Simple, basic, important information
pharmacies, medical supplies
calls for assistance 4 weeks ago... Source: Haklay 2010 Buffer created around high quality object (Ordnance Survey MasterMap Integrated Transport Network (ITN)) accuracy +/- 1m
4 random 25 sq.km. blocks in London
Buffer widths: Motorway - 8m, A-roads - 5.6m, B-roads - 3.75m
Percentage of tested object that falls within the buffer is evaluated
328km Overlap > 80% (1 in 4 roads >95%, 85.80% South London)
Average positional error trends 11m... 5m for 1... 13 contributors Thackery St Dickens St Station St Data generally accurate where present
Inaccurate roads Factors affecting completeness:
Increase in contributor density tends to greater accuracy Walking Papers First 13 contributors to an area 1 sq.km. in size considerably increase that dataset's positional accuracy Source: Haklay 2010 Source: Haklay 2010 Attributed: Girres and Touya (2010) <1 contributor per sq.km => n contributions
1<3 contributors per sq.km => n x 5...10 contributions
3< contributors per sq.km => n x 10...100 contributions
Competitive? Contributor notices change
Again dependent on no of contributors
Natural Disasters eqviewer.co.nz
... + other sites Majority of OSM roads accurate
A number missing from OSM
1 service road out by >50m "No names" view Tag Info Chatham Islands - first area for which LINZ upload to OSM has been completed caveat Curated feeds Why curate? It can't be that accurate can it? It can't be that accurate can it? Crowd Sourced Data Leigh Hunt
http://www.splicegroup.co.nz Questions? Source: Flickr map view (Annual week long event held in Black Rock Desert, Northern Nevada) Compares various OSM renderings to Google renderings http://bit.ly/eyZPLh- ITO World - http://www.itoworld.com/product/osm/
http://bit.ly/h9hn7l - eq.org.nz launch post - http://blog.ushahidi.com/index.php/2011/02/24/launching-eq-org-nz-for-the-new-zealand-earthquake/
http://bit.ly/i38UIX - Geofabrik Map Compare - http://tools.geofabrik.de/mc/
http://bit.ly/giNgpL - OSM History- http://labs.geofabrik.de/history/
http://bit.ly/hYgyag - Taginfo - http://taginfo.openstreetmap.de/
http://bit.ly/hEpZ6y - OSM Qulaity Aussrance - http://wiki.openstreetmap.org/wiki/Quality_Assurance
http://bit.ly/i1Efbg - ITO Map - http://wiki.openstreetmap.org/wiki/ITO_Map ITO OSM Mapper Thanks (aerial photography, vector data) Napier City Council
Porirua City Council Geofabrik Map Compare See also:
OSM History ITO Map Soft-launched on 18 March 2011 for the UK and parts of Europe and also Bay Area in California.
Coverage will be increased during the rest of March/April. Lots of tools, of varying functionality, status and ease of use.
But there are some easier, ready to use tools... OSM Quality Assurance http://wiki.openstreetmap.org/wiki/Quality_Assurance