“The Semantic Web is not a separate Web but an extension of the current one, in which information is given well-defined meaning, better enabling computers and people to work in cooperation.” - this statement of Tim Berners-Lee has gained even more relevance since the start of this century.
The humanity is rapidly developing and persistently experiencing local and global challenges, such as global warming/climate change, dis-balances in demand and supply, among many others. Mastering most (if not all) of them require a behavior change. Behavioral change is difficult to achieve per se, and it is important that technology – as a major enabler - has a positive rather than a negative impact here.
Further, the dramatic growth of data volumes (Big Data, Internet of Things) and the data’s increased power and impact and on the people's daily lives are calling for new types, practices and policies of behavior with data.
These factors made the role of semantic technology even more crucial: in terms of providing a well-defined meaning, and eventually delivering Smart Data for a functional and fair data value chain.
Addressing the behavioural change with Smart Data, I discuss potential ICT solutions investigating the domain of energy efficient buildings. Particularly, our completed OpenFridge experiment will be presented: design and development of the Internet of Things data system with semantic and data analytics enablers for building new services on a top of typical home appliance data — in particular, refrigerators. The system has been evaluated with real life end-user pilots.
In conclusions, I overview our related ongoing work, namely, in the areas of the impact of Big Data on society and related research roadmapping (linking to sociology), personalized energy efficiency data management services in buildings (linking to psychology), and semantic data licensing (linking to law).
Smart Data for Behavioural Change: Towards Energy Efficient Buildings
1. SMART DATA FOR BEHAVIOURAL CHANG
E:
TOWARDS ENERGY EFFICIENT BUILDINGS
Anna Fensel
Semantic Technology Institute (STI) Innsbruck, University of Innsbruck, Austria
Contact: anna.fensel@sti2.at
23.03.2017, Lunch Seminar of Institute of Computer Science, University Innsbruck, Austria
2. Motivation
Background
Main Story: OpenFridge
Extensions / Ongoing further work
- ENTROPY project: linking to psychology
- BYTE project and BBDC: linking to sociology
- DALICC project: linking to law
Outline
4. From: “The Semantic Web is not a separate
Web but an extension of the current one, in
which information is given well-defined
meaning, better enabling computers and people
to work in cooperation.” – Tim Berners-Lee, James
Hendler, Ora Lassila, 2001
Till: Smart Data??
Solved Problem
5. Going mainstream:
schema.org,...
Linked Open Data cloud
counts 25 billion triples
Open government initiatives
BBC, Facebook, Google,
Yahoo, etc. use semantics
SPARQL becomes W3C
recommendation
Life science and other
scientific communities use
ontologies
RDF, OWL become W3C
recommedations
Research field on ontologies
and semantics appears
Term „Semantic Web“ has
been „seeded“, Scientific
American article, Tim
Semantic Web Evolution in One
Slide
2008
2001
2010
2004 Source: Open Knowledge Foundation
12. Making Smart Data from Big Data with
semantics for energy efficient buildings -
Where it started…
Image credit: kurier.at
13. In the topic since 2009, with 5 national and EU funded projects
Was present in media such as:
Graduated several PhD, Master and Bachelor students
Published and reviewed at high quality energy venues e.g. Energy
Efficiency journal (SCI IF 2015: 1.183), but also at high quality Computer
Science venues
Received awards e.g. Highly Commended Paper 2015 of Int. J. of
Pervasive Computing and Communications, Best Short Paper Award
of iiWAS 2013
Gave invited talks e.g. at Skolkovo, ESTC
My credentials in the topic of energy
efficiency, smart buildings, responsible use
of associated data
15. School in Upper
Austria
Factory floor in Russia
Real Smart Building Setups
16. Smart Building Installations
Motivation: work with real buildings, real data
and real users
Technology:
Several Smart Meters
Sensors (e.g. light, temperature, humidity)
Smart plugs, for individual sockets
Multi-utility management
(i.e. electricity, heating)
Shutdown services for PCs
User interfaces and apps: Web, tablet,
smartphone (Android)
17. Data-Driven Management in the
Intelligent Building - SESAME-S Project
- Millions of real life data triples collected
in a semantic repository
- Ontology published at CKAN
18. Services Addressing Users @ School
Energy awareness,
monitoring
Remote control - manual and
programmed - e.g. scheduled activities
(ON/OFF policies) and triggering rules
(Alert sending rules)
How do we get the users?
By having workshops with pupils:
introduction to energy efficiency,
building analysis, explaining the
system and services
+ building administrators
20. Fensel, A., Tomic, S.D.K., Koller, A. “Contributing to
Appliances’ Energy Efficiency with Internet of Things, Smart
Data and User Engagement”. Future Generation Computer
Systems, Elsevier.
DOI:
SCI-indexed journal, 2015 Impact Factor: 2.430; CORE journal
rank: A
• Fensel, A., Gasser, F., Mayr, C., Ott, L., & Sarigianni, C. (2014). Selecting
Ontologies and Publishing Data of Electrical Appliances: A Refrigerator
Example. In On the Move to Meaningful Internet Systems: OTM 2014 Workshops
(pp. 494-503). Springer.
The presentation is based on…
http://dx.doi.org/10.1016/j.future.2016.11.026
21. Smart Grid is a Showcase for Data Economy
Smart Grid
Operation
Energy Markets
Synchro
Phasers
Renewables
Parks
Compliance
Smart Buildings
Electro
Mobility
Smart Cities
Smart
Appliances
Smart
Metering
Plant
Automation
Business
DSM
Compliance
Price Signals
Demand
Response
Capacity
Management
Prosumers
From general project presentation: http://www.slideshare.net/slotomic/big-data-openfrid
22. What is energy efficiency?
– Using less energy to provide equivalent
service.
– A life-cycle characteristic of home
appliances.
Economy for Energy Efficiency Data
(Knowledge)?
How energy efficiency is being assessed?
– By measuring and comparison.
– EE of Design: Efficiency labels awarded by
– verification institutes.
– EE of Use: Best practices, comparisons.
How potential for increasing energy efficiency
is being assessed?
– By measuring/comparison More context
needed
More info: http://www.atlete.eu, http://eetd.lbl.gov/ee/ee-1.htmlFrom general project presentation: http://www.slideshare.net/slotomic/big-data-openfrid
23. Metering (Data)
- A source of big data, two-way exchange
- Dynamic tariffs, distributed generation, demand
management
- Granularity of measurements aggregated vs. appliance
level
- Provides energy awareness context
A Value-chain for Energy Efficiency Data
Energy Awareness (Knowledge)
- Awareness context vs. usage context
- Awareness at the energy service level needed.
- Smart-plugs for individual measurements
- Label is a decision support tool pointing to technological
improvements in energy efficiency of appliances.
Efficiency Increasing Actions
- Appliance replacement, more efficient use, technology
improvements From general project presentation: http://www.slideshare.net/slotomic/big-data-openfrid
24. Developing a crowdsourcing platform for data collection
Exploring the concept of context-dependent energy
efficiency
Combining (big) data and semantics for add-value
services
OpenFridge : Opening and Processing
Appliances Data for Energy Efficiency
Improved
labeling
Improved
technology
and CRM
Better
decisions
about
replacement
and use
Home Users
Labeling Institutions
Manufacturers
Energy
Efficiency
Data
Building an ecosystem around data
From general project presentation: http://www.slideshare.net/slotomic/big-data-openfrid
25. Usage profile
avg.
consumption,
cooling cycle,
defrost
cycle,…
Appliance profile
type, volume,
producer,
efficiency,
year of
production,
stand-
alone/built-in,
facing south,
location:
kitchen / cellar,
city, country,
number of users
Measurement profile
cooling level (1,2,3,..),
inside temperature, room
temperature, level of
filling,
doors opening events,
measurement duration
Comparisons, Recommendations & Analytics Services
Compare different refrigerators, refrigerators of the same type,
performance at different environmental conditions, set-points and
loadings, impact of opening the door, of aging, of installation, …
From Context to Recommendations
Measurements
power level (5s)
timestamp
From general project presentation: http://www.slideshare.net/slotomic/big-da
26. Hardware & service interfaces for data acquisition
- Currently based on the existing commercial system with web-
service interface
Big data & analytics for data processing
- Anticipating large user base
Semantic technology for value-add services
- Easy integration of external data, vocabularies and ontologies
from the ecommerce and energy efficiency domain
- Logic-based reasoning
Privacy and security protection of data
- Data provenance and veracity
Community building and crowdsourcing
- Incentives based on high-quality recommendations
Platform Enablers
From general project presentation: http://www.slideshare.net/slotomic/big-data-openfrid
27. Interfaces
- Attractiveness and usability of user interfaces for data acquisition
- Instrumentation for appliances data acquisition
- Privacy of user and appliances data
- Accuracy of data
Big Data
- Analytics on raw data: mappers/reducers feed semantic
knowledgebase with model data
Semantic Layer
- Ontology engineering
- External data integration
- Performance of the semantic knowledgebase
- Expressiveness of services via SPARQL queries for B2B/B2C
portal-based analytics
Challenges
From general project presentation: http://www.slideshare.net/slotomic/big-data-openfrid
28. Community &
Content Management
Big Data
Infrastructure
Data Acquisition
Web Service
Drupal Portal &
Web Service Client
Recommendations
&
Visualizations
Appliance Profile
Measurements
Profile
Appliance Profile
Measurements
Profile
Measurements
Business
Intelligence
Services
Users
Manufacturers
Labeling
Organisations
OpenFridge Architecture
Semantic
Knowledg
e
Base
Analytics
SPARQL: Data-
as-a-Service
Usage Profile
Volume?
Variety?
Velocity?
Veracity?
Value?
From general project presentation: http://www.slideshare.net/slotomic/big-d
34. Tool: Python
• Importation process
• Restructure process
• Creation of the ontology-file
Result:
• OpenFridge ontology published at: http://www.sti-
innsbruck.at/results/ontologies, and indexed at LOV
portal
• 1032 refrigerator models with 18665 triples
Data Mapping – Implementation & Results
35. Technical:
● How to design an ontology 100% reusing other schemes
● How to fetch data from different HTML Web sources
● Screen scraping tools
● Creation of readable instances in protege
● How to get this data into a format that is readalbe for a tool like
Protege
○ How to develop
○ Challenges
Organizational:
● Managing project (devide tasks)
● Meetings (how to communicate)
● Engagement
Lessons Learned
36. Actions
- Interactions with the users
- Instrumentation @Home
- Privacy & data quality
Data (Big Data)
- Efficient storage
- Analytic processing, data structures
Semantic Processing
- Ontology Design
- Integration of external data from structured and non-
structured sources
- Development and optimisation of queries (SPARQL)
for added value servies
User Tests
- Project partner internal
- With test users & external
Implementation Steps
OpenFridge@WFF, Oct 2013
From general project presentation: http://www.slideshare.net/slotomic/big-data-openfrid
37. Our Goal: A platform for crowdsourcing of energy efficiency
data and a community for propagation of energy efficiency
social values
Exploring the concept of context-dependent energy
efficiency:
- Measurements in a broader context of different usage parameters
within a community of users
- Providing necessary explanations to motivate corresponding users’
actions towards improving the energy efficiency of services
Integrating Big Data and semantic technology
- Maintaining large volumes of raw data, analytics to transform raw
data into the parameterized information
- Developing appropriate ontologies to link parameterized energy
efficiency information with the usage context information
Developing semantic-based delivery of add-value services
- Querying and reasoning
Focusing on refrigerators as they are the largest energy
consuming home appliance; the same principles could be
further extended
Summary of the OpenFridge Platform
From general project presentation: http://www.slideshare.net/slotomic/big-data-openfrid
42. Aims:
To identify whether the users were capable of using the
platform as a whole, and their response rates to it, on the
hardware, software and service levels
- between October 2014 and September 2015, the platform acquired 68
active users. Engaged were ca. 100, but for the rest the platform did
not work for different reasons (hardware failure, wireless
incompatibility, inability to set-up)
To receive the feedback on the system’s existing and potential
features, particularly, regarding actual and potential usage of
collected data
- Survey at the end: 21 respondents (19 male, 2 female),
- 66,7%, came from Austria – the rest from the rest of the world,
- 90,5% of responders were running OpenFridge on Windows operation
system, and the rest were split between Linux, Android, iPhone/iPad.
Evaluation and Results
43. What do you see as the most useful
feature(s) and impact of the OpenFridge
portal?
44. Which data and knowledge engineering
issues on the portal have you experienced?
45. Would you share the data collected from
your appliances, and under which
conditions?
46. Which features do you think would increase
your engagement with the platform?
47. An Internet of Things semantic platform OpenFridge is designed and
implemented.
The platform has been deployed and evaluated with globally-
distributed real-life users.
Real-life user and fridge measurements data has been collected and
published in open source repositories.
A set of selected characteristic anonymous fridge and freezer measurements,
including the detailed observation data of 426 complete cooling cycles, composition
of used ontologies, data of >1032 refrigerator models:
in the datahub: http://datahub.io/dataset/the-measurement-data-set-from-the-project-
open-fridge, http://www.sti-innsbruck.at/results/ontologies, and indexed at LOV portal
High potential in facilitation of data economy has been
demonstrated in evaluations.
Challenges in deployment of such platforms are discussed.
Summary - Highlights
49. • The H2020 ENTROPY project aims to design and deploy an innovative IT
ecosystem targeting at improving energy efficiency through consumers
understanding, engagement and behavioural changes.
• http://entropy-project.eu
• 3 real-life pilots (Italy, Spain, Switzerland)
• Energy efficiency facilitation taking into account personality profiles of
the users
• Ongoing work
50. Energy beliefs
• 97% believe energy conservation is something to be concerned
about
• 95% feel that conserving energy and natural resources is important
to them
• 94% believe conserving energy also their problem
• 91% have responsibility to conserve energy and resources
• 90% believe the organization they work for should conserve energy
• 92% believe they should and would help organization conserve
energy
• >85% willing to change daily routine to conserve energy
however…
• Only 55% agree their country is in the middle of an energy crisis
• 20% feel news reports about energy crisis are blown out of
proportion
• >70% believe it is their right to use as much energy as they want
Innsbruck - Austria
2-3 February, 2017
51. Energy behaviour
• Almost all turn off the room/bathroom lights when
they leave
• 70% turn off computers
• >60% turn off the PC monitor
• ~50% turn off Air Conditioner(s)
• 23% turn off printer(s)
• 14% often leave the windows open with Aircon on
Innsbruck - Austria
2-3 February, 2017
52. Gamification User Types
• Using player typologies to understand individual
preferences is one of the common approaches for
personalization
• Personalizing gameful systems more effective than one-
size-fits-all approaches.
• Several studies indicated the need for personalizing
gamified systems to users’ personalities.
• Personalization can be used in gameful design to tailor
interaction mechanics to the user.
Innsbruck - Austria
2-3 February, 2017
53. HEXAD Gamification - User Types 1-3
Hexad gamification user types (Tondello et al., 2016):
• Philanthropists – motivated by purpose, altruistic and willing to give
without expecting a reward.
• Suggested design elements: collection and trading, gifting, knowledge sharing,
and administrative roles.
• Socialisers – motivated by relatedness – want to interact with others
and create social connections.
• Suggested design elements: guilds or teams, social networks, social
comparison, social competition, and social discovery.
• Free Spirits – motivated by autonomy, freedom to express themselves
and act without external control – like to create and explore within a
system.
• Suggested design elements: exploratory tasks, nonlinear gameplay, Easter
eggs, unlockable content, creativity tools, and customization.
Innsbruck - Austria
2-3 February, 2017
54. HEXAD Gamification - User Types 3-6
• Achievers – motivated by competence – seek to progress within a
system by completing tasks, or prove themselves by tackling difficult
challenges.
• Suggested design elements: challenges, certificates, learning new skills,
quests, levels or progression, and epic challenges (or “boss battles”).
• Players – motivated by extrinsic rewards – will do everything to earn a
reward within a system, independently of the type of the activity.
• Suggested design elements: points, rewards or prizes, leaderboards, badges or
achievements, virtual economy, and lotteries or games of chance.
• Disruptors – motivated by triggering changes – tend to disrupt the
system either directly or through others to force negative or positive
changes, test the system’s boundaries and try to push further. Although
disruption can be negative (e.g., cheaters or griefers), it can also work
towards improving the system.
• Suggested design elements: innovation platforms, voting mechanisms,
development tools, anonymity, anarchic gameplay.
Innsbruck - Austria
2-3 February, 2017
Although users are likely to display a principal tendency, in most cases
they will also be motivated by all the other types to some degree
(Tondello et al., 2016).
55. Gamification user types
• Achiever rated high by 89% of participants.
• Philanthropist by 88%
• Socializer by 76%
• Free Spirit by 75%
• Player by 43%
• Disruptor by 12%
Innsbruck - Austria
2-3 February, 2017
56. Correlation of user types & game elements
In addition to gamification user type prefs offered in
bibliography, for an energy conservation app, our sample
prefer:
• Philanthropists badges and roles.
• Socialisers points, badges, rewards and roles.
• Free spirits points, badges, progression, status, levels and
roles.
• Achievers no specific preference towards any of the
elements.
• Disruptors status.
• Players rewards, points, badges, leaderboards, status
Innsbruck - Austria
2-3 February, 2017
57. Personality Profile
– The big 5 personality traits have been
correlated with:
– Pro-environmental attitudes & environmental
engagement
– Concern For Privacy in LBS & Usage Intention of
Location-Based Services
– Game Playing Style, Behaviour, Motivations to
Play, Difficulty adaptation
– Player typologies
– Game genre preferences
Innsbruck - Austria
2-3 February, 2017
58. Engagement
• The “positive work-related state of fulfilment that is
characterized by vigor, dedication, and absorption”, the
positive antipode of burnout. (Schaufeli, Bakker and
Salanova, 2006)
• Gallup’s categorization of employees, based on level of
engagement (Prakash and Rao, 2015):
• Engaged: work with passion and feel a profound connection to
their organization, drive innovation and move the organization
forward
• Non-engaged: are essentially “checked-out”, sleepwalking through
their workday, putting time but not energy or passion into their
work
• Actively disengaged: are not just unhappy at work, but busy acting
out their unhappiness, undermining what their colleagues
accomplish, every day
Innsbruck - Austria
2-3 February, 2017
59. BYTE:
The BYTE research roadmap
Anna Fensel and Marti Cuquet,
University of Innsbruck, Austria
BYTE final conference, London, UK, 9 February 2017
Big data roadmap and cross-disciplinary community for
addressing societal externalities
60. Starting points: research topics from
BDVA and literature survey
• Research topics from BDVA’s Strategic Research and Innovation Agenda.
• Defines overall goals, technical and non-technical priorities and a research and innovation
roadmap.
• 6 main priorities:
Data
management
Data
processing
Data
analytics
Data
protection
Data
visualisation
Non-technical
priorities
to handle
unstructured data,
ensure semantic
interoperability,
asses data quality
and provenance
Optimised and
efficient
architectures for
data-at-rest and
data-in-motion,
decentralised,
scalable
with improved
models and
simulations, semantic
analysis, pattern
discovery, business
intelligence and
predictive and
prescriptive analytics
and anonymisation
to enable not open
data enter the Data
Value Chain with a
complete data
protection
framework,anonymis
ation algorithms,
multiparty mining
and user experience,
with interactive and
personalised
visualisations,
simplified query and
discovery
mechanisms, linked
data visualisations
skills development,
standardisation,
social perceptions
and societal
implication.
61. Data management Data processing Data analytics Data protection Data visualisation
Non-technical
priorities
A1 Handling
unstructured data
B1 Architectures for
data-at-rest and
data-in-motion
C1 Improved models
and simulations
D1 Complete data
protection
framework
E1 End user
visualisation and
analytics
F1 Establish and
increase trust
A2 Semantic
interoperability
B2 Tools for processing
real-time
heterogeneous
data
C2 Semantic analysis D2 Data minimization E2 Dynamic clustering
of information
F2 Privacy-by-design
A3 Measuring and
assuring data
quality
B3 Scalable algorithms
and techniques for
real-time analytics
C3 Event and pattern
discovery
D3 Privacy-preserving
mining algorithms
E3 New visualisation
for geospatial data
F3 Ethical issues
A4 Data management
lifecycle
B4 Decentralised
architectures
C4 Multimedia
(unstructured) data
mining
D4 Robust
anonymisation
algorithms
E4 Interrelated data
and semantics
relationships
F4 Develop new
business models
A5 Data provenance,
control and IPR
B5 Efficient
mechanisms for
storage and
processing
C5 Deep learning
techniques for BI,
predictive and
prescriptive
analytics
D5 Protection against
reversibility
E5 Qualitative analysis
at a high semantic
level
F5 Citizen research
A6 Data-as-a-service
model and
paradigm
C6 Context-aware
analytics
D6 Pattern hiding
mechanism
E6 Real-time and
collaborative 3-D
visualisation
F6 Discrimination
discovery and
prevention
D7 Secure multiparty
mining mechanism
E7 Time dimension of
big data
E8 Real-time adaptable
and interactive
visualisation
62. Process
1) Discussion and validation of
research topics
•Work in small round tables.
• Are the topics representative?
• Are there other relevant topics or subtopics?
• Are there other relevant sources aside from SRIA you’d like to incorporate?
2) Alignment of research topics
and externalities
• BYTE identified externalities have been grouped in 4 groups
and 18 subgroups
3) Time alignment and prioritisation
65. BYTE Big Data Research Roadmap -
Summary
• Presents positive and negative externalities of big data in 18 industry sectors.
• Maps research to its societal impact and contribution to skills and standards.
• Provides a timeline for research efforts with its impact on each sector.
• Summarises best practices to capture the positive societal benefits of big data.
• Compact version: Cuquet, M., & Fensel, A. (2016). Big data impact on society:
a research roadmap for Europe. arXiv preprint arXiv:1610.06766. URI:
https://arxiv.org/abs/1610.06766
• Full version as D6.1 BYTE deliverable: http://byte-project.eu/research
• Join BYTE Big Data Community (BBDC): http://new.byte-project.eu/byte-
community
66. Data licensing
Image from DALICC consortium: FH St Pölten, STI Innsbruck, WU Wien,
Semantic Web Company, Höhne i. d. Maur & Partner Rechtsanwälte OG
https://www.dalicc.net/
Data licensing is still complicated, formats for licensed data use are under-defined.
Semantic standards for license development are in progress e.g. ODRL, RightsML.
Automated semantic-based data licensing support for derivative works is our ongoing work.
Statement
More complex and elaborate network
Non trivial problems
Smart grid,
Objectives and business model
Stakeholders
Generate data
Rules how these
Generator of big data
Operation on the consumer side
Smart Cities
Markets
Consumer side is quite interesting – new business models – control of energy
Smart metering smart appliances
Energy efficiency
Is there an economy for Energy efficiency Data?
What data
Three questions
First question
Using less for equivalent
Life-cycle question
Diagram
Measurements –
Let us look at the value chain
All starts with the measurements – data
Based on data new knowledge can be created
And then some actions can be undertaken to increase energy efficiency
So it all starts with the meter, the smart meter,
We have been talking about Smart meters for years
They can measure and communicate
The support dynamic tariffs, distributed generation and
What am I aware of
Awareness context – how much we used – usage context much granular
OpenFridge is a research project funded by the Austrian research funding agency