What would you do with free pictures of everything on Earth? Paul Houle
Jul 07, 2015
What would you do with free pictures of everything on Earth?
Paul Houle
How it all started
Photo Credit: Daniel Sparing The slides to this talk are licensed CC-BY-SA/3.0
Flickr kills the dream…
But Creative Commons Saves the Day
animalphotos.info
dbpedia& the semantic web
<dbpedia:Albatross><rdf:type><dbpedia-owl:Bird> .
Subject Predicate (Verb) Object
http://dbpedia.org/resource/Albatross
http://www.w3.org/1999/02/22-rdf-syntax-ns#type
http://dbpedia.org/ontology/Bird
Taxonomy Construction and Data Integration
Output
Triple Store RDF File CMS NoSQL Lucene
RDF Processing Pipeline
OWL SPARQL 1.1 SPIN RIF Hadoop
Input Data
RDF Freebase Relational CSV
automating the process
dbpedia flickr
Identify topics search for candidates filter correct images
describe images
amazon mechanical turk
carpictures.cc
Fueleconomy.gov
2007 2008 2009
Chevrolet Honda Volkswagen
Civic ElementAccord
Wikipedia
Honda Concept Cars
Front-Wheel Drive Vehicles
S360 FCX Civic
Chevrolet Honda Volkswagen
Civic ElementAccordS360 FCX
the business model
People use images
People make links
Site gets more traffic
Advertising Revenue
Gather content
how many things exist?
The game of 20 questions 220 = about a million…
what sort of things exist?
“Everything on Earth:” Places, People, Creative Works, Life Forms, Technological Artifacts
ny-pictures.com
geospatial selection + Wikipedia graph
The only way is no way…
The only limits are no limits…
The only taxonomy is no taxonomy…
exploring the earth’s noösphere
Noösphere, according to the thought of Vladimir Vernadsky and Teilhard de Chardin, denotes the "sphere of human thought".
people
placesinventions
creative works
life forms
ookaboo.com takes off
ookaboo
pictures
topics
sources
Network analysis+
Text analysis+
Social
ookaboo
ookaboo semantic API
http://en.wikipedia.org/wiki/Thailand
API
Thanks: Andyindia, Echiner1, Rene Eherhardt
Why don’t computers understand human language?
The Ink is In The Pen
The Pig is In The PenThanks: Seaniz, GiangHồThịHoàng, Claudia Scholz and
the human memome project
One ring to bind them all?
Top-down? Bottom-up? Both!
Merge data from a wide range of sources, continuously test against reality by performance in practical applications
Databases Logic
21st century knowledge engineering
RDF
“machine learning” + rules + social + …
A Market in Common Sense
Linked Data Business Models
Free Shared Vocabulary Enables Interconnection…
…but the profit motive spurs investment to create quality data.
Ookaboo directions
• Official RDF dump
• More pictures
• Better navigation
• Better accuracy
• Semantically targeted advertising
Questions?
http://ookaboo.com/
All images on Ookaboo are public domain or creative commons and can be used freely for commercial and non-commercial purposes