Building intelligent systems with FAIR data

BUILDING INTELLIGENT SYSTEMS
WITH F.A.I.R. DATA
Ilaria Tiddi
Faculty of Computer Science && Faculty of Behavioural Sciences
Vrije Universiteit Amsterdam
@IlaTiddi

F.A.I.R. DATA ARE JUST PRINCIPLES
Make your data
● F → Findable (through persistent URIs)
● A → Accessible (in trustable repositories)
● I → Interoperable (using the same formalisms)
● R → Reusable (through usage licences)

BUT WHAT CAN I DO WITH IT?
Allow intelligent systems to find data:
● F → Serendipitously
● A → Cross-domain
● I → Automatically
● R → Trustworthy

INTELLIGENT SYSTEMS THAT CAN EXPLAIN
My case: intelligent systems that explain to humans:
● facts
● decisions
● behaviours
Explaining : generating coherence between old and new knowledge

INTELLIGENT SYSTEMS THAT CAN EXPLAIN
Why do we need systems generating explanations?
● to learn new knowledge
● to find meaning (reconciling contradictions in our knowledge)
● to socially interact (creating a shared meaning with the others)
● ...and because GDPR says so
Users have a “right to explanation”
for any decisions made about them

USING F.A.I.R. DATA TO EXPLAIN
Assumption: Linked (F.A.I.R.) Data can serve as background knowledge to
generate explanations:
● Plenty of available sources
● Connected, centralised hubs
● Multi-domain datasets, allowing serendipity

[1] Tiddi. (2016), Explaining Data Patterns using Knowledge from the Web of Data, Ph.D. thesis. Demo: http://dedalo.kmi.open.ac.uk/
EXPLAINING DATA PATTERNS
Why do people search for
“A Song of Ice and Fire”
only in certain periods?”

[1] Tiddi. (2016), Explaining Data Patterns using Knowledge from the Web of Data, Ph.D. thesis. Demo: http://dedalo.kmi.open.ac.uk/
EXPLAINING DATA PATTERNS
Why do women in the
yellow countries have a
lower literacy rate?

EXPLAINING BEHAVIOURS
Explaining behaviours to
self-learners with DBpedia
and providing Open University
course recommendations
[2] http://afel-project.eu
[3] http://data.open.ac.uk

EXPLAINING DATASET BIAS
Use DBpedia to discover:
● The NYT dataset is about
places in the US (trivial)
● The Reading Experience Dataset
is about poets/novelists which
committed suicide (less trivial)
[4] Tiddi. (2014), Quantifying the bias in data links (EKAW2014)
owl:sameAs
skos:exactMatch
...
A
B
Projection of B in A

EXPLAINING RADIO CONTENTS
Augmenting smart-city
applications with Linked Data
(DBpedia, MK:DataHub)
[5] Tiddi et al. (2018), Allowing exploratory search from podcasts: the case of Secklow Sounds Radio (ISWC2018)
[6] http://datahub.mksmart.org

EXPLAINING SCENES IN MOTION
Object classification using
Interconnected knowledge bases
DBpedia ConceptNet ShapeNet

EXPLAINING NEURAL ATTENTIONS
Explaining multi-layer LSTM
networks using linguistic
corpora (e.g. FrameNet)
[7] Mensio et al., A Multi-layer LSTM-based Approach for Robot Command Interaction Modeling, Language and Robotics (LangRobo), IROS 2018.
[8] https://framenet.icsi.berkeley.edu/fndrupal/

EXPLAINING HUMAN COOPERATION
Cooperation Databank : 50 years
of studies on human cooperation
Using Linked Data to support
scientific knowledge discovery

EXPLAINING MACHINE ETHICS
Bringing social and computer
scientists together
Reflect on the threats and misuse
of current technologies
[8] https://kmitd.github.io/recoding-black-mirror/

Thank you
...and all of them!
@IlaTiddi
i.tiddi@vu.nl

Building intelligent systems with FAIR data

More Related Content

What's hot (20)

Similar to Building intelligent systems with FAIR data (20)

More from Vrije Universiteit Amsterdam (11)

Recently uploaded (20)

Building intelligent systems with FAIR data