histograph for historians
DESCRIPTION
2014, 24 March, Insa de Lyon, France, presentation of the CUbRIK histoGraph, by Lars Wieneke (CVCE)TRANSCRIPT
The CVCE
1 Lars Wieneke, CVCE, Luxembourg
About CUbRIK
n European Community's Seventh Framework Program FP7-ICT
n 15 European partners n Multimedia search
processing: Putting humans in the loop
n Demos: History of Europe and Fashion
2 Lars Wieneke, CVCE, Luxembourg
Point of departure
Images as sources
3 Lars Wieneke, CVCE, Luxembourg
4 Lars Wieneke, CVCE, Luxembourg
Goal: Reconstructing and exploring social ties through historical sources
5 Lars Wieneke, CVCE, Luxembourg
Towards the social graph 4 pillars
1. Close connection to the requirements ofresearchers in European Integration studies
2. Structured and referencable repository ofpersons, events and places in time
3. Efficient indexation process that enables the association of faces with identities
4. Toolchain for analysis and visualization
6 Lars Wieneke, CVCE, Luxembourg
Towards the social graph Sourcing researcher requirements
Selection of target user group
First draft of the app scenario
Feedback on technical scope
Exploratory interviews
(daily work practices)Second draft of the
app scenario
Focus group(user needs and app
scenarios) Feedback on technical feasability
Lessons learned:issues and features
Specification
Implementation 1. demonstrator
Workshop: Review of app and features
Revised specification
Implementation 2. demonstrator
Evaluation and test
Stage 1
Stage 2
Stage 3
Stage 4
Stage 5
Users
Requirements
Technology
7 Lars Wieneke, CVCE, Luxembourg
Towards the social graph Structured repository
Towards the social graph
Indexation process
8
Raw content
High level features (automatic annotations)
Conflict (e.g., “Image contains
‘Romano Prodi’ ” Confidence = low) ?
Conflict store Conflict manager
Conflict resolution task store
Conflict resolution task: conflict,
required skill, priority, ..
CUbRIK app for Conflict resolution
Game Q&A Crowdtask
Lars Wieneke, CVCE, Luxembourg
9
Towards the social graph Indexation process II
n Human in the loop added value: n Verification of identities/places/events ambiguous and temporal only
possible by putting humans in the loop n Integration of multiple perspectives
n CUbRIK as an open toolbox allows follow-up and extension through third parties
n “Vertical” integration: GUI, components, crowdsourcing
integrated in a platform
Lars Wieneke, CVCE, Luxembourg
10
Towards the social graph Visualization and analysis
Lars Wieneke, CVCE, Luxembourg
11
Challenges & Approach
n Main challenges n Detection and identification of identities/places/events in time n Verification of identities/places/events in time n Analysis of relationships (e.g. co-occurrences) n Rights aware crawling and storage n Verification of provenance and license information n Truth and provenance
n Approach n Crowd-sourced verification of detected faces (false positives/negatives) n Verification of identities through/places/events in time social networks of
experts n Visual knowledge discovery/exploration n Integrated rights aware crawling and storage n Integrated license and provenance management
Lars Wieneke, CVCE, Luxembourg
Towards the social graph Bringing it all together
12 Lars Wieneke, CVCE, Luxembourg
Image Indexation
Media harvesting and upload
Face detection
Face identification
Clickworkers
Crowd Face position
validation
Copyright aware
crawler
Provenance checker
License checker
Content provider
tools
Metadata Entity
extraction
Identity reconciliation
Entity verification & annotation
Entitypedia Integration
CROWD pre-
filtering
Text Indexation
Connection to the CVCE collection
Entity anntation and
extraction
Expert Crowd
Expert CROWD
verification
Entitypedia Integration
CROWD Research Inquieries
Expert Crowd
CROWD Research Inquiry
Social Graph Network Analysis
Graph Visualization
Analysis of the social
graph
Graph Query (old: Query for Entities)
Graph Visualization
Query for entities
Context Expander
Expansion through
documents
Expansion through videos
Expansion through images
Expansion through related entities
Social Graph construction
Social Graph
Creation
Content Analysis and Enrichment Querying Feedback acquisition
and processing Y3 component
Graph Visualization
Query for spatial
constraints
EXP through SIMILAR images
WP Event detection
histoGraph demo
Time for a Demo!
13 Lars Wieneke, CVCE, Luxembourg
Pipelining the CUbRIK components: Human input from click-workers
Great choice for simple tasks: n Face detection: false positives, false negatives n Monetary motivation, via www.microtask.com
Poor performance on complex tasks: n Low resolution images n Different angles etc. n Actors recurring over time
14 Lars Wieneke, CVCE, Luxembourg
Pipelining the CUbRIK components: Human input from experts
Capable of complex tasks: n In-depth knowledge of key actors n Context knowledge allows inferences
But: Different motivational models! n Public goods n Reputation
15 Lars Wieneke, CVCE, Luxembourg
Usage for historians
n No one truth in history but interpretation, context and discussion
n Therefore need to represent ambivalence, contradictions and discussion
n Close ties between data representation (Social graph) and their original context (primary sources)
16 Lars Wieneke, CVCE, Luxembourg
Conclusion
n Challenges n What is truth? Humanities vs. Computer Science n Gathering requirements for tools that haven‘t been
developed yet n Engaging crowds n Image copyrights n Scientific value?
n Refinement of the application n Additional datasources n Improvement of the interface n Integration of the new components
17 Lars Wieneke, CVCE, Luxembourg
Outlook
18
Outlook
19
Outlook histoGraph
20
WWW.CUBRIKPROJECT.EU
Visit us on
21
@CUBRIKPROJECT
Or follow us on Twitter