Data-intensive decision making in the era of big data and artificial intelligence

Prof. Dr. Maria A. Wimmer
University of Koblenz-Landau, Germany
www.uni-Koblenz.de/agvinf
wimmer@uni-koblenz.de
Data-intensive decision making in the era of
big data and artificial intelligence
Revolutionising digital governance???

Objectives of presentation
 Meta-view: Contextualising data-intensive decision-making in public
sector
 Where can AI and Big Data be leveraged in decision-making
 What future research and training needs can be identified
(insights from Gov 3.0)
2020/07/14 2Samos Summit 2020, (c) Prof. Dr. Maria A. Wimmer

Scope of decision-making in public spheres
Public Policy Making
Law-Making
Public Service
Evaluation &
Impact Assessment

Contextualizing decision-making in public spheres
Decision-making along public policy making
[Policy Lifecycle by Howlett, M., & Ramesh, M.
Studying Public Policy: Policy Cycles and Policy
Subsystems. Toronto: Oxford University Press, 1995]

Data as an
asset
AI and Big Data supporting Policy Decision-making
[Policy Lifecycle by Howett and Ramesh, 1995
Data Asset Management Lifecycle: https://www.lean-data.nl/data-as-an-asset/]
Analysis of Social Media for Opinion Mining, Sentiment Analysis and Agenda setting
Text Mining for Policy making
Mining Open Data for Policy making
Leveraging structured and unstructured data …
Data assets for data-driven policy analysis and
modelling, supporting policy formulation and decision-
making (e.g. using simulations of various kinds,
gamification etc.), …
AI-based support in policy decision-making
Assessing and simulating different policy options and their consequences
through AI and policy simulation
…

Decision-making along Law-making pcrocedures
Second
reading in
Council
Second
reading in
Parliament
First reading
in CouncilFirst reading
in Parliament
Commission
proposal
Conciliation
[https://www.europarl.europa.eu/infographic/legislative-procedure/index_en.html]
Third
reading
in the
Parliament
and Council

AI and Big Data supporting Law-making
AI and Big Data Analytics for better informed Law-making
AI and Big Data for ex-ante legal impact assessment
Gamification in consultation and deliberation on draft bills
Lots of
data
Lots of
opinions
Lots of
findings
from studies,
monitoring,
consultations
…
Lots of
Expert
Knowledge
AI and ML
Big Data Analytics
Simulations & Gamification

Decision-making in Implementation &
Execution of Policy or Legislation
Executive
Bodies
Public Administration
Public Service
Provisioning
Citizens
Companies, NGOs
Observation &
Monitoring
§

Areas where AI and Big Data can support Policy Implementation
and Implementation of Legislation
AI and Robotics in automatic public service
execution
Robot process automation
Expert systems supporting public servants in
decisions along public service
AI and Big data in civil protection and crime prevention
(surveillance and cybercrime), …
AI and Big data support in real-time interventions in
society and market
Big data analytics in monitoring the evolution of different
policy domains: environment, health, climate, growth,
finance and austerity, education, etc.

Improved Government
and Governance
Impact of
Public
Service
Impact of
Legislation
Impact of
Public
Policy
Intervention
Needs from
Monitoring/
Observation
Ad-hoc
Decision-
making Needs
(e.g. pandemic
outbreaks)
…
Decision-making in Evaluation and Revisions
AI and ML
Big Data Analytics
Simulations &
Gamification
Dashboards

Areas where AI and Big Data can support
Evaluation and Revisions of Policy / Legislation
Automatic analysis and monitoring of performance of public service (in
different policy domains) for better informed decision-making
Analysis and simulation for ex-post policy impact assessment and ex-post
legal impact assessment
Big Data Analytics applied to [open and closed] Data repositories on
people, companies, environment and other policy areas to identify needed
policy interventions
Dashboards for societal and market evolution, for monitoring the evolution
in relevant policy areas, and to support better public policy-making and law-
making
Data … Data … Data … Data

Examples of AI use in Government:
EC‘s AI Watch on Artificial Intelligence in Public Services
[https://ec.eur
opa.eu/jrc/en/
publication/ai-
watch-2019-
activity-report,
p. 18]

Gov 3.0 – Roadmap for Research
and Training Needs in
Government 3.0
https://www.gov30.eu/

Roadmapping method
Roadmap
Research
gaps &
needs
Training
gaps &
needs
Scenarios
Internet of things and Smart cities
Artificial intelligence and
machine learning
Data driven policy modelling
Virtual and augmented reality
Natural language processing &
sentiment analysis
Etc.
Blockchain
Projectanalysisandsynthesisinvolving
disruptivetechnologiesindigital
government
Insightsfromthedescriptionofdisruptive
technologies
Step 1 Step 2 Step 3 Step 4
[Wimmer, Viale Pereira, Ronzhyn, Spitzer (2020). Transforming Government by Leveraging
Disruptive Technologies. In eJournal of eDemocracy and Open Government. 12 (1): 87-114]

Scenario examples
 Included possible future implementations in AI, ML, NLP, IoT, AR/VR and
Blockchain technologies
 Implementations of smart city, gamification & co-creation of public
services with the support of particular disruptive technologies

Workshop approach to identify
research & training needs
04.09.2019
Discussing scenarios with disruptive technologies:
Identifying research and training needs:

Research needs on disruptive
technologies in Government 3.0
AI/ML
BigData
IoT
Gamification
AR/VR
NLP
Blockchain
Cloud(fog)Computing
eID/eSignature
SmartCity
Co-creation
CommunityAwarenes
Platforms
Once-onlyPrinciple
Open(Linked)
GovernmentData
ServiceModules
Gaming-basedPolicy
Modellingand
Simulation
Standardisation and interoperability
of disruptive technologies
    
Analysis of stakeholders                
Evaluation and policy making     
Data security and data privacy      
Automated decision-making  
Ethical issues   
Disruptive Technologies Concepts of Government 3.0 using disrupt. t.

Training needs on disruptive
technologies in Government 3.0
AI/ML
BigData
IoT
Gamification
AR/VR
NLP
Blockchain
CloudComputing
eID/eSignature
SmartCity
Co-creation
CommunityAwarenes
Platforms
Once-onlyPrinciple
Open(Linked)
GovernmentData
ServiceModules
Gaming-basedPolicy
Modellingand
Simulation
General technology skills      
New technologies in public
management & digital
government
  
Management and economics
capabilities on the use of
disruptive technologies
               
Capabilities in data science, data
security and legal compliance
          
Capabilities in responsible
research and in sustainability
               
Disruptive Technologies Concepts of Government 3.0 using disrupt. t.

Challenges and outlook
 AI and Big Data bear big potentials for revolutionising the way of public
sector service towards more evidence-based data-driven decision-making
 How these potentials can be leveraged effectively, and legally and socially
compliant, still needs substantial interdisciplinary research efforts
 For understanding the most influential phenomena of risks and success
 For designing and implementing solutions that meet the needs and expectations
of society
 For understanding how the solutions work in practice and create impact

Challenges and outlook
 AI Watch of JRC (EC) and a number of newly emerging case studies in
research provide good insights
 Multi-faceted challenges to overcome (see research and training needs
identified in Gov 3.0)
 Motivation to change, fear of being exposed to manipulation (algorithms)
and misuse (data), and acceptance by people to be tackled (social issue)
 Particular trade-off between exploiting machine intelligence and at the same
time meeting the social needs of people
Good balance between technological advancements and the pace of socially
compatible digital transformation demands for interdisciplinary research

Many thanks for your attention!
wimmer@uni-koblenz.de
http://www.uni-koblenz.de/agvinf

With the use of AI and Big Data in policy-making, law-making
and public service provisioning, will the ways of decision-
making substantially change in the next 1-2 decades ?
https://pingo.coactum.de/517165
PIN: 517165
Question for discussion:

Results from online query: Question 1

If your previous answer tended towards change, please describe what will change in decision-making by using AI and Big Data (Nr.
of entries=17, usable answers by individuals= 16, nr. of answers analysed=26, multiple choice):
Time to decisions Soft factors Input to decision-making
Way of decision-
making
Results of
decision-making
Governance
model
Faster decision-
making (3
occurences)
Better meeting
needs and
problems;
Increased equality;
More transparency
Easier decisions;
Handling data will improve;
Impact assessment;
More evidence based on expertise;
More informed drafting of
regulation;
More real-world data;
More sentiment analysis;
New information sources used in
decision-making and policy-making;
New source of legitimization of
politicians' choices;
Suggestions of most probable
courses of action
Automated
decision-making
will be more
prevalent
Less decisions
based on intuition
or ideology;
Outcomes of
decisions
will change;
Quicker evaluation
and monitoring;
Results more
reliable
Interaction with
surroundings;
Interconnectivity;
Less hierarchy;
More flexibility;
Public servants'
workforce will be
restructured

If your previous answer tended towards NO change, please explain why you think that decision-making
procedures will not change by using AI and Big Data (8 participants answering):
 No change in political decisions in the past 20 years / Tradition
 Resistance of politicians to leave decisions to AI and big data / Resistance among public servants
 Policy makers will not take decision-making with AI and big data seriously
 Need to distinguish linear thinking and exponential thinking: if people will be able to learn to think
more exponentially, we will have more parts changing because of AI
 Data not being harmonized
 Organisational and/or cultural hesitation
 Interoperability of systems
 Ethical concerns

Data-intensive decision making in the era of big data and artificial intelligence

More Related Content

What's hot (20)

Similar to Data-intensive decision making in the era of big data and artificial intelligence (20)

More from samossummit (20)

Recently uploaded (20)

Data-intensive decision making in the era of big data and artificial intelligence