SlideShare a Scribd company logo
Learning Pulse
D. Di Mitri, M. Scheffel, H. Drachsler, D. Börner, S. Ternier, M. Specht
A machine learning approach for
predicting performance in self-regulated
learning using multimodal data
Paper presentation at LAK17
15th March 2017, Vancouver, Canada
Outline
1. Background, context, vision
2. Our approach
3. Data collection
4. Data analysis
5. Conclusions
Pagina 2
Data deluge in education
Pagina 3
Collecting learning experiences
Picture from tincanapi.com
Pagina 4
Pagina 5
Learning happening across spaces
Context: Self Regulated Learning
Self-Regulated Learning → no guidance → no feedback → no support
Pagina 6
Vision: machine learning approach
y = f(X)
Learning
Performance
(output space)
Predictive
Model
Multimodal Data
(input space)
Pagina 7
Our approach
Pagina 8
Research questions
(RQ-MAIN) How can we store, model and analyse
multimodal data to predict performance in human learning?
(RQ1) Which architecture allows the collection and storage
of multimodal data in a scalable and efficient way?
(RQ2) What is the best way to model multimodal data to
apply supervise machine learning techniques?
(RQ3) Which machine learning model is able to produce
learner specific predictions on multimodal data?
Pagina 9
Participants
• 9 PhD students at Welten institute
• Different disciplines
• Different working setups:
– Time
– Tasks
– Operating systems
Pagina 10
Experimental timeline
Pagina 11
Phase 0
Pre-test
System architecture tested
Phase 1
Training
3 weeks of data collection
Phase 2
Validation
2 weeks of data collection and prediction
Input space – multimodal data
Pagina 12
Context
Body
Activities
Body: physiological (heart-rate)
and physical responses
(steps) - from Fitbit HR
Activities: applications used
during learning
from RescueTime
Context: weather data
from OpenWeatherMap
Output space – Flow Csikszentmihalyi, 1972
Pagina 13
Theoretical Empirical
Activity Rating Tool
Productivity
How productive was
last activity?
Stress
How stressful was
last activity?
Challenge
How challenging was
last activity?
Abilities
How prepared did you
feel for the activity?
FLOW
Participants rate hourly, from 7AM to 7PM
A scalable web app!
Client: Bootstrap + Jquery
Sever: GoogleApp + Python
“Very easy to
use!”
Pagina 14
Data collection
Pagina 15
Data model
Pagina 16
Berg, A., Scheffel, M., Drachsler, H., Ternier, S. & Specht, M. (2016). The Dutch xAPI Experience. Proceedings of the 6th
International Conference on Learning Analytics and Knowledge (LAK’16), April 25-29, 2016, Edinburgh, UK.
Data storing format for the Learning Record StoreExperience API
The data journey
Pagina 18
Complex architecture
Pagina 19
Data collection
• PULL data from the 3rd party APIs
• Make the xAPI triples
• PUSH data in the LRS
• It’s scalable!
• No collisions
• It’s fast
• It’s Interoperable
Learning Pulse Server
+
Learning Record Store
Pagina 20
Data Processing Application
Script in Python running on a VM which processes data in real time
Pagina 21
Data Analysis
Pagina 22
Transformed dataset
• Time Series: tabular representation
• 5 minutes intervals
• Enough samples now!
• Easier view for Machine Learning
• Signal resampling needed
9410
observations
X
29 attributes
Pagina 23
Issue 1) Feature extraction from Time Series
Heart Rate Variability and
Heart Rate Entropy…
didn’t work
SOLUTION
• Mean of the signal
• Maximum
• Minimum
• Standard Deviation
• Average change
Heart-ratesignalfor15mins
Pagina 24
Issue 2) Activity data very sparse
Rule based grouping of applications
Learners’ activity can be compared!
Applications used are
too sparse
SOLUTION
Let’s create
application categories
Pagina 25
Issue 3) Number of labels available
Trade-off:
number of labels
vs
Seamlessness of the data
collections
NO SOLUTION
Pagina 26
Issue 5) Random vs continuous data
Independence constraint
Knowing one value of et for
one observation does not
help us to guess value of
et+1
yt = α + βX t + et
cov(et ,et+1) = 0
FIXED Effect
RANDOM Effect
SOLUTION follows...
Pagina 27
Mixed Effect Linear Model
x0 x1 x2
... xn-1 xn g y
t0 x x x ... x x 1 y
t1 x x x ... x x 1 y
t2 x x x ... x x 2 y
t... ... ... ... ... ... ... 2 y
tp
-1
x x x x x x 3 y
tp ? ? --- --- --- --- x ?
Random EffectsFixed Effects Group
Used R-squared for
goodness-test
LIMITATIONS
● Convergence time
● Mono-output
Pagina 28
Issue 6) Inter-subject variability
i.e. Participants have
rated very differently
SOLUTION
Predictions are
normalised wrt each
learner
xnew = (xmax – xmin)
*xi/100 + xmin
Pagina 29
Conclusions
Pagina 30
RQ1) Architecture
The architecture developed was able of:
1. Importing great number of sensor data in xAPI
format;
2. combining sensor data with self-reports
3. programmatically transform xAPI data
4. train predictive models & reuse them
5. save the predictions to compare with actual values
Pagina 31
RQ2) Represent multimodal data
• Multiple Instance Representation
• Each learning sample is a 5 minute interval
• It’s suitable for machine learning
Pagina 32
RQ3) Machine learning model
• Linear Mixed Effect Models allow
1. taking into account data specific to each
learner
2. distinguish between fixed and random
effects
3. Take categorical data into account.
Pagina 33
Limitations
• Low accuracy of predictions
R-Square tests Stress: 0.32, Challenge: 0.22, Flow score:
0.16, Abilities: 0.08, Productivity: 0.05.
• Real-time issues
Fitbit synchronisation, Virtual Machine performance
• 3rd party API constraints
• No great solution for grouping activity data (manual
grouping)
Pagina 34
Opportunities
• Data driven
• Real Time feedback
• Visualisations can show feedback
• Seamless data collection
• Multimodal dataset for reserach
• Reusable architecture
Pagina 35
*Börner, Tabuenca, Storm,
Happe, and Specht. 2015
Example visualisation:
The Feedback Cube*
Q&A
Thanks for listening!
Daniele Di Mitri
ddm@ou.nl
@dimstudi0
Pagina 36
Check my poster!

More Related Content

What's hot

2015 03 19 (EDUCON2015) eMadrid UAM Towards MOOCs scenaries based on Collabor...
2015 03 19 (EDUCON2015) eMadrid UAM Towards MOOCs scenaries based on Collabor...2015 03 19 (EDUCON2015) eMadrid UAM Towards MOOCs scenaries based on Collabor...
2015 03 19 (EDUCON2015) eMadrid UAM Towards MOOCs scenaries based on Collabor...
eMadrid network
 
Learner Ontological Model for Intelligent Virtual Collaborative Learning Envi...
Learner Ontological Model for Intelligent Virtual Collaborative Learning Envi...Learner Ontological Model for Intelligent Virtual Collaborative Learning Envi...
Learner Ontological Model for Intelligent Virtual Collaborative Learning Envi...
ijceronline
 
2015 03 19 (EDUCON2015) eMadrid UC3M Reports from eMadrid Network about Blend...
2015 03 19 (EDUCON2015) eMadrid UC3M Reports from eMadrid Network about Blend...2015 03 19 (EDUCON2015) eMadrid UC3M Reports from eMadrid Network about Blend...
2015 03 19 (EDUCON2015) eMadrid UC3M Reports from eMadrid Network about Blend...
eMadrid network
 
Teaching about AR and Teaching with AR
Teaching about AR and Teaching with ARTeaching about AR and Teaching with AR
Teaching about AR and Teaching with AR
Förderverein Technische Fakultät
 
Experimenting multiple approaches for teaching meta-modeling
Experimenting multiple approaches for teaching meta-modelingExperimenting multiple approaches for teaching meta-modeling
Experimenting multiple approaches for teaching meta-modeling
Saïd Assar
 
Designing and eXperiencing Smart Objects based Learning Scenarios: an approac...
Designing and eXperiencing Smart Objects based Learning Scenarios: an approac...Designing and eXperiencing Smart Objects based Learning Scenarios: an approac...
Designing and eXperiencing Smart Objects based Learning Scenarios: an approac...
Technological Ecosystems for Enhancing Multiculturality
 
2015 03 19 (EDUCON2015) eMadrid UPM Towards a Learning Analytics Approach for...
2015 03 19 (EDUCON2015) eMadrid UPM Towards a Learning Analytics Approach for...2015 03 19 (EDUCON2015) eMadrid UPM Towards a Learning Analytics Approach for...
2015 03 19 (EDUCON2015) eMadrid UPM Towards a Learning Analytics Approach for...
eMadrid network
 
IEEE EDUCON 2015 reputation mooc
 IEEE EDUCON  2015 reputation mooc IEEE EDUCON  2015 reputation mooc
IEEE EDUCON 2015 reputation mooc
Miguel R. Artacho
 
Applying Soft Computing Techniques in Information Retrieval
Applying Soft Computing Techniques in Information RetrievalApplying Soft Computing Techniques in Information Retrieval
Applying Soft Computing Techniques in Information Retrieval
IJAEMSJORNAL
 
The Future for Technologies in Schools: FutureSchools 2017
The Future for Technologies in Schools: FutureSchools 2017The Future for Technologies in Schools: FutureSchools 2017
The Future for Technologies in Schools: FutureSchools 2017
Michael A. Cowling
 
Towards blended reality on collaborative laboratory activities using smart ob...
Towards blended reality on collaborative laboratory activities using smart ob...Towards blended reality on collaborative laboratory activities using smart ob...
Towards blended reality on collaborative laboratory activities using smart ob...
Anasol Pena-Rios
 
Deep Learning and CNN Architectures
Deep Learning and CNN ArchitecturesDeep Learning and CNN Architectures
Deep Learning and CNN Architectures
Ferdin Joe John Joseph PhD
 
Deep Learning Projects - Anomaly Detection Using Deep Learning
Deep Learning Projects - Anomaly Detection Using Deep LearningDeep Learning Projects - Anomaly Detection Using Deep Learning
Deep Learning Projects - Anomaly Detection Using Deep Learning
DezyreAcademy
 
Ppt tale kn_intro_final
Ppt tale kn_intro_finalPpt tale kn_intro_final
Ppt tale kn_intro_final
Manuel Castro
 
Do Higher Education Institutions Need a Digital Passport
Do Higher Education Institutions Need a Digital PassportDo Higher Education Institutions Need a Digital Passport
Do Higher Education Institutions Need a Digital Passport
The Mind Lab
 
Mobile Scenarios Presentation
Mobile Scenarios PresentationMobile Scenarios Presentation
Mobile Scenarios Presentation
Math4Mobile
 
Piloting Mixed Reality in ICT Networking to Visualize Complex Theoretical Mul...
Piloting Mixed Reality in ICT Networking to Visualize Complex Theoretical Mul...Piloting Mixed Reality in ICT Networking to Visualize Complex Theoretical Mul...
Piloting Mixed Reality in ICT Networking to Visualize Complex Theoretical Mul...
Bond University
 
Aalto CCIS Programme
Aalto CCIS ProgrammeAalto CCIS Programme
Aalto CCIS Programme
Petri Vuorimaa
 
A Hidden Markov Model Approach to Predict Students' Actions in an Adaptive an...
A Hidden Markov Model Approach to Predict Students' Actions in an Adaptive an...A Hidden Markov Model Approach to Predict Students' Actions in an Adaptive an...
A Hidden Markov Model Approach to Predict Students' Actions in an Adaptive an...
Simon Bolivar University
 

What's hot (20)

2015 03 19 (EDUCON2015) eMadrid UAM Towards MOOCs scenaries based on Collabor...
2015 03 19 (EDUCON2015) eMadrid UAM Towards MOOCs scenaries based on Collabor...2015 03 19 (EDUCON2015) eMadrid UAM Towards MOOCs scenaries based on Collabor...
2015 03 19 (EDUCON2015) eMadrid UAM Towards MOOCs scenaries based on Collabor...
 
Learner Ontological Model for Intelligent Virtual Collaborative Learning Envi...
Learner Ontological Model for Intelligent Virtual Collaborative Learning Envi...Learner Ontological Model for Intelligent Virtual Collaborative Learning Envi...
Learner Ontological Model for Intelligent Virtual Collaborative Learning Envi...
 
2015 03 19 (EDUCON2015) eMadrid UC3M Reports from eMadrid Network about Blend...
2015 03 19 (EDUCON2015) eMadrid UC3M Reports from eMadrid Network about Blend...2015 03 19 (EDUCON2015) eMadrid UC3M Reports from eMadrid Network about Blend...
2015 03 19 (EDUCON2015) eMadrid UC3M Reports from eMadrid Network about Blend...
 
Teaching about AR and Teaching with AR
Teaching about AR and Teaching with ARTeaching about AR and Teaching with AR
Teaching about AR and Teaching with AR
 
Experimenting multiple approaches for teaching meta-modeling
Experimenting multiple approaches for teaching meta-modelingExperimenting multiple approaches for teaching meta-modeling
Experimenting multiple approaches for teaching meta-modeling
 
Designing and eXperiencing Smart Objects based Learning Scenarios: an approac...
Designing and eXperiencing Smart Objects based Learning Scenarios: an approac...Designing and eXperiencing Smart Objects based Learning Scenarios: an approac...
Designing and eXperiencing Smart Objects based Learning Scenarios: an approac...
 
2015 03 19 (EDUCON2015) eMadrid UPM Towards a Learning Analytics Approach for...
2015 03 19 (EDUCON2015) eMadrid UPM Towards a Learning Analytics Approach for...2015 03 19 (EDUCON2015) eMadrid UPM Towards a Learning Analytics Approach for...
2015 03 19 (EDUCON2015) eMadrid UPM Towards a Learning Analytics Approach for...
 
IEEE EDUCON 2015 reputation mooc
 IEEE EDUCON  2015 reputation mooc IEEE EDUCON  2015 reputation mooc
IEEE EDUCON 2015 reputation mooc
 
Inaugural lecture
Inaugural lectureInaugural lecture
Inaugural lecture
 
Applying Soft Computing Techniques in Information Retrieval
Applying Soft Computing Techniques in Information RetrievalApplying Soft Computing Techniques in Information Retrieval
Applying Soft Computing Techniques in Information Retrieval
 
The Future for Technologies in Schools: FutureSchools 2017
The Future for Technologies in Schools: FutureSchools 2017The Future for Technologies in Schools: FutureSchools 2017
The Future for Technologies in Schools: FutureSchools 2017
 
Towards blended reality on collaborative laboratory activities using smart ob...
Towards blended reality on collaborative laboratory activities using smart ob...Towards blended reality on collaborative laboratory activities using smart ob...
Towards blended reality on collaborative laboratory activities using smart ob...
 
Deep Learning and CNN Architectures
Deep Learning and CNN ArchitecturesDeep Learning and CNN Architectures
Deep Learning and CNN Architectures
 
Deep Learning Projects - Anomaly Detection Using Deep Learning
Deep Learning Projects - Anomaly Detection Using Deep LearningDeep Learning Projects - Anomaly Detection Using Deep Learning
Deep Learning Projects - Anomaly Detection Using Deep Learning
 
Ppt tale kn_intro_final
Ppt tale kn_intro_finalPpt tale kn_intro_final
Ppt tale kn_intro_final
 
Do Higher Education Institutions Need a Digital Passport
Do Higher Education Institutions Need a Digital PassportDo Higher Education Institutions Need a Digital Passport
Do Higher Education Institutions Need a Digital Passport
 
Mobile Scenarios Presentation
Mobile Scenarios PresentationMobile Scenarios Presentation
Mobile Scenarios Presentation
 
Piloting Mixed Reality in ICT Networking to Visualize Complex Theoretical Mul...
Piloting Mixed Reality in ICT Networking to Visualize Complex Theoretical Mul...Piloting Mixed Reality in ICT Networking to Visualize Complex Theoretical Mul...
Piloting Mixed Reality in ICT Networking to Visualize Complex Theoretical Mul...
 
Aalto CCIS Programme
Aalto CCIS ProgrammeAalto CCIS Programme
Aalto CCIS Programme
 
A Hidden Markov Model Approach to Predict Students' Actions in an Adaptive an...
A Hidden Markov Model Approach to Predict Students' Actions in an Adaptive an...A Hidden Markov Model Approach to Predict Students' Actions in an Adaptive an...
A Hidden Markov Model Approach to Predict Students' Actions in an Adaptive an...
 

Viewers also liked

A Search-based Testing Approach for XML Injection Vulnerabilities in Web Appl...
A Search-based Testing Approach for XML Injection Vulnerabilities in Web Appl...A Search-based Testing Approach for XML Injection Vulnerabilities in Web Appl...
A Search-based Testing Approach for XML Injection Vulnerabilities in Web Appl...
Lionel Briand
 
Modernismo americano.
Modernismo americano. Modernismo americano.
Modernismo americano.
valeriarahal
 
TCI network for practitioners
TCI network for practitionersTCI network for practitioners
TCI network for practitioners
Dr. Amit Kapoor
 
Actividad 11 de yamileth lópez requena
Actividad 11 de yamileth lópez requenaActividad 11 de yamileth lópez requena
Actividad 11 de yamileth lópez requena
Yamileth López Requena
 
Departamento de producción
Departamento de producciónDepartamento de producción
Departamento de producción
Monserrat Correa
 
Presentation on healthy relationships
Presentation on healthy  relationshipsPresentation on healthy  relationships
Presentation on healthy relationships
corrieperdok
 
CEO’ların sunum sırları
CEO’ların sunum sırlarıCEO’ların sunum sırları
CEO’ların sunum sırları
Sunumo
 
#stillhuman : how ethnography makes sense of data @sxsw2017
#stillhuman : how ethnography makes sense of data @sxsw2017 #stillhuman : how ethnography makes sense of data @sxsw2017
#stillhuman : how ethnography makes sense of data @sxsw2017
_unknowns
 
BITCOIN USER BASE DOUBLING EVERY 12 MONTHS: GOOGLE TRENDS
BITCOIN USER BASE DOUBLING EVERY 12 MONTHS: GOOGLE TRENDSBITCOIN USER BASE DOUBLING EVERY 12 MONTHS: GOOGLE TRENDS
BITCOIN USER BASE DOUBLING EVERY 12 MONTHS: GOOGLE TRENDS
Steven Rhyner
 
Dreamforce to you mx jan18
Dreamforce to you   mx jan18 Dreamforce to you   mx jan18
Dreamforce to you mx jan18
Salesforce Latinoamérica
 
Culture in banking is everything
Culture in banking is everythingCulture in banking is everything
Culture in banking is everything
The Retail Banking Academy
 

Viewers also liked (11)

A Search-based Testing Approach for XML Injection Vulnerabilities in Web Appl...
A Search-based Testing Approach for XML Injection Vulnerabilities in Web Appl...A Search-based Testing Approach for XML Injection Vulnerabilities in Web Appl...
A Search-based Testing Approach for XML Injection Vulnerabilities in Web Appl...
 
Modernismo americano.
Modernismo americano. Modernismo americano.
Modernismo americano.
 
TCI network for practitioners
TCI network for practitionersTCI network for practitioners
TCI network for practitioners
 
Actividad 11 de yamileth lópez requena
Actividad 11 de yamileth lópez requenaActividad 11 de yamileth lópez requena
Actividad 11 de yamileth lópez requena
 
Departamento de producción
Departamento de producciónDepartamento de producción
Departamento de producción
 
Presentation on healthy relationships
Presentation on healthy  relationshipsPresentation on healthy  relationships
Presentation on healthy relationships
 
CEO’ların sunum sırları
CEO’ların sunum sırlarıCEO’ların sunum sırları
CEO’ların sunum sırları
 
#stillhuman : how ethnography makes sense of data @sxsw2017
#stillhuman : how ethnography makes sense of data @sxsw2017 #stillhuman : how ethnography makes sense of data @sxsw2017
#stillhuman : how ethnography makes sense of data @sxsw2017
 
BITCOIN USER BASE DOUBLING EVERY 12 MONTHS: GOOGLE TRENDS
BITCOIN USER BASE DOUBLING EVERY 12 MONTHS: GOOGLE TRENDSBITCOIN USER BASE DOUBLING EVERY 12 MONTHS: GOOGLE TRENDS
BITCOIN USER BASE DOUBLING EVERY 12 MONTHS: GOOGLE TRENDS
 
Dreamforce to you mx jan18
Dreamforce to you   mx jan18 Dreamforce to you   mx jan18
Dreamforce to you mx jan18
 
Culture in banking is everything
Culture in banking is everythingCulture in banking is everything
Culture in banking is everything
 

Similar to Learning Pulse - paper presentation at LAK17

Parallel Filter-Based Feature Selection Based on Balanced Incomplete Block De...
Parallel Filter-Based Feature Selection Based on Balanced Incomplete Block De...Parallel Filter-Based Feature Selection Based on Balanced Incomplete Block De...
Parallel Filter-Based Feature Selection Based on Balanced Incomplete Block De...
AMIDST Toolbox
 
Deep Learning and Automatic Differentiation from Theano to PyTorch
Deep Learning and Automatic Differentiation from Theano to PyTorchDeep Learning and Automatic Differentiation from Theano to PyTorch
Deep Learning and Automatic Differentiation from Theano to PyTorch
inside-BigData.com
 
Learning from meaningful, purposive interaction
Learning from meaningful, purposive interactionLearning from meaningful, purposive interaction
Learning from meaningful, purposive interaction
fridolin.wild
 
230208 MLOps Getting from Good to Great.pptx
230208 MLOps Getting from Good to Great.pptx230208 MLOps Getting from Good to Great.pptx
230208 MLOps Getting from Good to Great.pptx
Arthur240715
 
Linked Data: Een extra ontstluitingslaag op archieven
Linked Data: Een extra ontstluitingslaag op archieven Linked Data: Een extra ontstluitingslaag op archieven
Linked Data: Een extra ontstluitingslaag op archieven
Richard Zijdeman
 
Modellbildung, Berechnung und Simulation in Forschung und Lehre
Modellbildung, Berechnung und Simulation in Forschung und LehreModellbildung, Berechnung und Simulation in Forschung und Lehre
Modellbildung, Berechnung und Simulation in Forschung und Lehre
Joachim Schlosser
 
M.Sc. Thesis Topics and Proposals @ Polimi Data Science Lab - 2024 - prof. Br...
M.Sc. Thesis Topics and Proposals @ Polimi Data Science Lab - 2024 - prof. Br...M.Sc. Thesis Topics and Proposals @ Polimi Data Science Lab - 2024 - prof. Br...
M.Sc. Thesis Topics and Proposals @ Polimi Data Science Lab - 2024 - prof. Br...
Marco Brambilla
 
Moving forward data centric sciences weaving AI, Big Data & HPC
Moving forward data centric sciences  weaving AI, Big Data & HPCMoving forward data centric sciences  weaving AI, Big Data & HPC
Moving forward data centric sciences weaving AI, Big Data & HPC
Genoveva Vargas-Solar
 
Linked Open Data: Combining Data for the Social Sciences and Humanities (and ...
Linked Open Data: Combining Data for the Social Sciences and Humanities (and ...Linked Open Data: Combining Data for the Social Sciences and Humanities (and ...
Linked Open Data: Combining Data for the Social Sciences and Humanities (and ...
Richard Zijdeman
 
(Explainable) Data-Centric AI: what are you explaininhg, and to whom?
(Explainable) Data-Centric AI: what are you explaininhg, and to whom?(Explainable) Data-Centric AI: what are you explaininhg, and to whom?
(Explainable) Data-Centric AI: what are you explaininhg, and to whom?
Paolo Missier
 
AI and Deep Learning
AI and Deep Learning AI and Deep Learning
AI and Deep Learning
Subrat Panda, PhD
 
2016 03-16 digital energy luncheon
2016 03-16 digital energy luncheon2016 03-16 digital energy luncheon
2016 03-16 digital energy luncheon
Mark Reynolds
 
Knowledge Graph Embeddings for Recommender Systems
Knowledge Graph Embeddings for Recommender SystemsKnowledge Graph Embeddings for Recommender Systems
Knowledge Graph Embeddings for Recommender Systems
Enrico Palumbo
 
Thoughts on Knowledge Graphs & Deeper Provenance
Thoughts on Knowledge Graphs  & Deeper ProvenanceThoughts on Knowledge Graphs  & Deeper Provenance
Thoughts on Knowledge Graphs & Deeper Provenance
Paul Groth
 
2016 05-20-clariah-wp4
2016 05-20-clariah-wp42016 05-20-clariah-wp4
2016 05-20-clariah-wp4
CLARIAH
 
Spark-MPI: Approaching the Fifth Paradigm with Nikolay Malitsky
Spark-MPI: Approaching the Fifth Paradigm with Nikolay MalitskySpark-MPI: Approaching the Fifth Paradigm with Nikolay Malitsky
Spark-MPI: Approaching the Fifth Paradigm with Nikolay Malitsky
Databricks
 
H2O with Erin LeDell at Portland R User Group
H2O with Erin LeDell at Portland R User GroupH2O with Erin LeDell at Portland R User Group
H2O with Erin LeDell at Portland R User Group
Sri Ambati
 
Himansu sahoo resume-ds
Himansu sahoo resume-dsHimansu sahoo resume-ds
Himansu sahoo resume-ds
Himansu Sahoo
 
YandongWang_Resume
YandongWang_ResumeYandongWang_Resume
YandongWang_Resumeyandong wang
 
Data legend dh_benelux_2017.key
Data legend dh_benelux_2017.keyData legend dh_benelux_2017.key
Data legend dh_benelux_2017.key
Richard Zijdeman
 

Similar to Learning Pulse - paper presentation at LAK17 (20)

Parallel Filter-Based Feature Selection Based on Balanced Incomplete Block De...
Parallel Filter-Based Feature Selection Based on Balanced Incomplete Block De...Parallel Filter-Based Feature Selection Based on Balanced Incomplete Block De...
Parallel Filter-Based Feature Selection Based on Balanced Incomplete Block De...
 
Deep Learning and Automatic Differentiation from Theano to PyTorch
Deep Learning and Automatic Differentiation from Theano to PyTorchDeep Learning and Automatic Differentiation from Theano to PyTorch
Deep Learning and Automatic Differentiation from Theano to PyTorch
 
Learning from meaningful, purposive interaction
Learning from meaningful, purposive interactionLearning from meaningful, purposive interaction
Learning from meaningful, purposive interaction
 
230208 MLOps Getting from Good to Great.pptx
230208 MLOps Getting from Good to Great.pptx230208 MLOps Getting from Good to Great.pptx
230208 MLOps Getting from Good to Great.pptx
 
Linked Data: Een extra ontstluitingslaag op archieven
Linked Data: Een extra ontstluitingslaag op archieven Linked Data: Een extra ontstluitingslaag op archieven
Linked Data: Een extra ontstluitingslaag op archieven
 
Modellbildung, Berechnung und Simulation in Forschung und Lehre
Modellbildung, Berechnung und Simulation in Forschung und LehreModellbildung, Berechnung und Simulation in Forschung und Lehre
Modellbildung, Berechnung und Simulation in Forschung und Lehre
 
M.Sc. Thesis Topics and Proposals @ Polimi Data Science Lab - 2024 - prof. Br...
M.Sc. Thesis Topics and Proposals @ Polimi Data Science Lab - 2024 - prof. Br...M.Sc. Thesis Topics and Proposals @ Polimi Data Science Lab - 2024 - prof. Br...
M.Sc. Thesis Topics and Proposals @ Polimi Data Science Lab - 2024 - prof. Br...
 
Moving forward data centric sciences weaving AI, Big Data & HPC
Moving forward data centric sciences  weaving AI, Big Data & HPCMoving forward data centric sciences  weaving AI, Big Data & HPC
Moving forward data centric sciences weaving AI, Big Data & HPC
 
Linked Open Data: Combining Data for the Social Sciences and Humanities (and ...
Linked Open Data: Combining Data for the Social Sciences and Humanities (and ...Linked Open Data: Combining Data for the Social Sciences and Humanities (and ...
Linked Open Data: Combining Data for the Social Sciences and Humanities (and ...
 
(Explainable) Data-Centric AI: what are you explaininhg, and to whom?
(Explainable) Data-Centric AI: what are you explaininhg, and to whom?(Explainable) Data-Centric AI: what are you explaininhg, and to whom?
(Explainable) Data-Centric AI: what are you explaininhg, and to whom?
 
AI and Deep Learning
AI and Deep Learning AI and Deep Learning
AI and Deep Learning
 
2016 03-16 digital energy luncheon
2016 03-16 digital energy luncheon2016 03-16 digital energy luncheon
2016 03-16 digital energy luncheon
 
Knowledge Graph Embeddings for Recommender Systems
Knowledge Graph Embeddings for Recommender SystemsKnowledge Graph Embeddings for Recommender Systems
Knowledge Graph Embeddings for Recommender Systems
 
Thoughts on Knowledge Graphs & Deeper Provenance
Thoughts on Knowledge Graphs  & Deeper ProvenanceThoughts on Knowledge Graphs  & Deeper Provenance
Thoughts on Knowledge Graphs & Deeper Provenance
 
2016 05-20-clariah-wp4
2016 05-20-clariah-wp42016 05-20-clariah-wp4
2016 05-20-clariah-wp4
 
Spark-MPI: Approaching the Fifth Paradigm with Nikolay Malitsky
Spark-MPI: Approaching the Fifth Paradigm with Nikolay MalitskySpark-MPI: Approaching the Fifth Paradigm with Nikolay Malitsky
Spark-MPI: Approaching the Fifth Paradigm with Nikolay Malitsky
 
H2O with Erin LeDell at Portland R User Group
H2O with Erin LeDell at Portland R User GroupH2O with Erin LeDell at Portland R User Group
H2O with Erin LeDell at Portland R User Group
 
Himansu sahoo resume-ds
Himansu sahoo resume-dsHimansu sahoo resume-ds
Himansu sahoo resume-ds
 
YandongWang_Resume
YandongWang_ResumeYandongWang_Resume
YandongWang_Resume
 
Data legend dh_benelux_2017.key
Data legend dh_benelux_2017.keyData legend dh_benelux_2017.key
Data legend dh_benelux_2017.key
 

More from Daniele Di Mitri

SenseTheClassroom Live at EC-TEL 2022
SenseTheClassroom Live at EC-TEL 2022SenseTheClassroom Live at EC-TEL 2022
SenseTheClassroom Live at EC-TEL 2022
Daniele Di Mitri
 
Guest Lecture: Restoring Context in Distance Learning with Artificial Intelli...
Guest Lecture: Restoring Context in Distance Learning with Artificial Intelli...Guest Lecture: Restoring Context in Distance Learning with Artificial Intelli...
Guest Lecture: Restoring Context in Distance Learning with Artificial Intelli...
Daniele Di Mitri
 
MOBIUS: Smart Mobility Tracking with Smartphone Sensors
MOBIUS: Smart Mobility Tracking with Smartphone SensorsMOBIUS: Smart Mobility Tracking with Smartphone Sensors
MOBIUS: Smart Mobility Tracking with Smartphone Sensors
Daniele Di Mitri
 
The Multimodal Tutor - short pitch presentation at JTELSS 2018 in Durrës, Alb...
The Multimodal Tutor - short pitch presentation at JTELSS 2018 in Durrës, Alb...The Multimodal Tutor - short pitch presentation at JTELSS 2018 in Durrës, Alb...
The Multimodal Tutor - short pitch presentation at JTELSS 2018 in Durrës, Alb...
Daniele Di Mitri
 
Sensors for Learning workshop
Sensors for Learning workshopSensors for Learning workshop
Sensors for Learning workshop
Daniele Di Mitri
 
Multimodal Machines #JTELSS17 workshop
Multimodal Machines #JTELSS17 workshopMultimodal Machines #JTELSS17 workshop
Multimodal Machines #JTELSS17 workshop
Daniele Di Mitri
 
Visual Learning Pulse
Visual Learning PulseVisual Learning Pulse
Visual Learning Pulse
Daniele Di Mitri
 
Digital Learning Projection - poster for #LAK17
Digital Learning Projection - poster for #LAK17Digital Learning Projection - poster for #LAK17
Digital Learning Projection - poster for #LAK17
Daniele Di Mitri
 
Digital Learning Projection - Learning state estimation from multimodal learn...
Digital Learning Projection - Learning state estimation from multimodal learn...Digital Learning Projection - Learning state estimation from multimodal learn...
Digital Learning Projection - Learning state estimation from multimodal learn...
Daniele Di Mitri
 
Academic writing in LaTeX
Academic writing in LaTeX Academic writing in LaTeX
Academic writing in LaTeX
Daniele Di Mitri
 
Visual Learning Pulse - Final Thesis presentation
Visual Learning Pulse - Final Thesis presentationVisual Learning Pulse - Final Thesis presentation
Visual Learning Pulse - Final Thesis presentation
Daniele Di Mitri
 
Research project MAI2 - Final Presentation Group 4
Research project MAI2  - Final Presentation Group 4Research project MAI2  - Final Presentation Group 4
Research project MAI2 - Final Presentation Group 4
Daniele Di Mitri
 
Lifelong Topic Modelling presentation
Lifelong Topic Modelling presentation Lifelong Topic Modelling presentation
Lifelong Topic Modelling presentation
Daniele Di Mitri
 
Word2Vec: Learning of word representations in a vector space - Di Mitri & Her...
Word2Vec: Learning of word representations in a vector space - Di Mitri & Her...Word2Vec: Learning of word representations in a vector space - Di Mitri & Her...
Word2Vec: Learning of word representations in a vector space - Di Mitri & Her...
Daniele Di Mitri
 
Battlecode2014 - final presentation - group n.2
Battlecode2014 - final presentation - group n.2Battlecode2014 - final presentation - group n.2
Battlecode2014 - final presentation - group n.2
Daniele Di Mitri
 
Inclusive & open learning analytics
Inclusive & open learning analyticsInclusive & open learning analytics
Inclusive & open learning analytics
Daniele Di Mitri
 
(IT) Slides della presentazione della tesi di Laurea
(IT) Slides della presentazione della tesi di Laurea(IT) Slides della presentazione della tesi di Laurea
(IT) Slides della presentazione della tesi di Laurea
Daniele Di Mitri
 
Claim Your Voice - Presentation of the results of the campign for VET students
Claim Your Voice - Presentation of the results of the campign for VET studentsClaim Your Voice - Presentation of the results of the campign for VET students
Claim Your Voice - Presentation of the results of the campign for VET students
Daniele Di Mitri
 
Obessu views on school resource
Obessu views on school resourceObessu views on school resource
Obessu views on school resource
Daniele Di Mitri
 
Obessu’s inputs on «opening up education»
Obessu’s inputs on «opening up education»Obessu’s inputs on «opening up education»
Obessu’s inputs on «opening up education»
Daniele Di Mitri
 

More from Daniele Di Mitri (20)

SenseTheClassroom Live at EC-TEL 2022
SenseTheClassroom Live at EC-TEL 2022SenseTheClassroom Live at EC-TEL 2022
SenseTheClassroom Live at EC-TEL 2022
 
Guest Lecture: Restoring Context in Distance Learning with Artificial Intelli...
Guest Lecture: Restoring Context in Distance Learning with Artificial Intelli...Guest Lecture: Restoring Context in Distance Learning with Artificial Intelli...
Guest Lecture: Restoring Context in Distance Learning with Artificial Intelli...
 
MOBIUS: Smart Mobility Tracking with Smartphone Sensors
MOBIUS: Smart Mobility Tracking with Smartphone SensorsMOBIUS: Smart Mobility Tracking with Smartphone Sensors
MOBIUS: Smart Mobility Tracking with Smartphone Sensors
 
The Multimodal Tutor - short pitch presentation at JTELSS 2018 in Durrës, Alb...
The Multimodal Tutor - short pitch presentation at JTELSS 2018 in Durrës, Alb...The Multimodal Tutor - short pitch presentation at JTELSS 2018 in Durrës, Alb...
The Multimodal Tutor - short pitch presentation at JTELSS 2018 in Durrës, Alb...
 
Sensors for Learning workshop
Sensors for Learning workshopSensors for Learning workshop
Sensors for Learning workshop
 
Multimodal Machines #JTELSS17 workshop
Multimodal Machines #JTELSS17 workshopMultimodal Machines #JTELSS17 workshop
Multimodal Machines #JTELSS17 workshop
 
Visual Learning Pulse
Visual Learning PulseVisual Learning Pulse
Visual Learning Pulse
 
Digital Learning Projection - poster for #LAK17
Digital Learning Projection - poster for #LAK17Digital Learning Projection - poster for #LAK17
Digital Learning Projection - poster for #LAK17
 
Digital Learning Projection - Learning state estimation from multimodal learn...
Digital Learning Projection - Learning state estimation from multimodal learn...Digital Learning Projection - Learning state estimation from multimodal learn...
Digital Learning Projection - Learning state estimation from multimodal learn...
 
Academic writing in LaTeX
Academic writing in LaTeX Academic writing in LaTeX
Academic writing in LaTeX
 
Visual Learning Pulse - Final Thesis presentation
Visual Learning Pulse - Final Thesis presentationVisual Learning Pulse - Final Thesis presentation
Visual Learning Pulse - Final Thesis presentation
 
Research project MAI2 - Final Presentation Group 4
Research project MAI2  - Final Presentation Group 4Research project MAI2  - Final Presentation Group 4
Research project MAI2 - Final Presentation Group 4
 
Lifelong Topic Modelling presentation
Lifelong Topic Modelling presentation Lifelong Topic Modelling presentation
Lifelong Topic Modelling presentation
 
Word2Vec: Learning of word representations in a vector space - Di Mitri & Her...
Word2Vec: Learning of word representations in a vector space - Di Mitri & Her...Word2Vec: Learning of word representations in a vector space - Di Mitri & Her...
Word2Vec: Learning of word representations in a vector space - Di Mitri & Her...
 
Battlecode2014 - final presentation - group n.2
Battlecode2014 - final presentation - group n.2Battlecode2014 - final presentation - group n.2
Battlecode2014 - final presentation - group n.2
 
Inclusive & open learning analytics
Inclusive & open learning analyticsInclusive & open learning analytics
Inclusive & open learning analytics
 
(IT) Slides della presentazione della tesi di Laurea
(IT) Slides della presentazione della tesi di Laurea(IT) Slides della presentazione della tesi di Laurea
(IT) Slides della presentazione della tesi di Laurea
 
Claim Your Voice - Presentation of the results of the campign for VET students
Claim Your Voice - Presentation of the results of the campign for VET studentsClaim Your Voice - Presentation of the results of the campign for VET students
Claim Your Voice - Presentation of the results of the campign for VET students
 
Obessu views on school resource
Obessu views on school resourceObessu views on school resource
Obessu views on school resource
 
Obessu’s inputs on «opening up education»
Obessu’s inputs on «opening up education»Obessu’s inputs on «opening up education»
Obessu’s inputs on «opening up education»
 

Recently uploaded

Thesis Statement for students diagnonsed withADHD.ppt
Thesis Statement for students diagnonsed withADHD.pptThesis Statement for students diagnonsed withADHD.ppt
Thesis Statement for students diagnonsed withADHD.ppt
EverAndrsGuerraGuerr
 
The approach at University of Liverpool.pptx
The approach at University of Liverpool.pptxThe approach at University of Liverpool.pptx
The approach at University of Liverpool.pptx
Jisc
 
Acetabularia Information For Class 9 .docx
Acetabularia Information For Class 9  .docxAcetabularia Information For Class 9  .docx
Acetabularia Information For Class 9 .docx
vaibhavrinwa19
 
Introduction to AI for Nonprofits with Tapp Network
Introduction to AI for Nonprofits with Tapp NetworkIntroduction to AI for Nonprofits with Tapp Network
Introduction to AI for Nonprofits with Tapp Network
TechSoup
 
Unit 8 - Information and Communication Technology (Paper I).pdf
Unit 8 - Information and Communication Technology (Paper I).pdfUnit 8 - Information and Communication Technology (Paper I).pdf
Unit 8 - Information and Communication Technology (Paper I).pdf
Thiyagu K
 
special B.ed 2nd year old paper_20240531.pdf
special B.ed 2nd year old paper_20240531.pdfspecial B.ed 2nd year old paper_20240531.pdf
special B.ed 2nd year old paper_20240531.pdf
Special education needs
 
Azure Interview Questions and Answers PDF By ScholarHat
Azure Interview Questions and Answers PDF By ScholarHatAzure Interview Questions and Answers PDF By ScholarHat
Azure Interview Questions and Answers PDF By ScholarHat
Scholarhat
 
Chapter 3 - Islamic Banking Products and Services.pptx
Chapter 3 - Islamic Banking Products and Services.pptxChapter 3 - Islamic Banking Products and Services.pptx
Chapter 3 - Islamic Banking Products and Services.pptx
Mohd Adib Abd Muin, Senior Lecturer at Universiti Utara Malaysia
 
Supporting (UKRI) OA monographs at Salford.pptx
Supporting (UKRI) OA monographs at Salford.pptxSupporting (UKRI) OA monographs at Salford.pptx
Supporting (UKRI) OA monographs at Salford.pptx
Jisc
 
Marketing internship report file for MBA
Marketing internship report file for MBAMarketing internship report file for MBA
Marketing internship report file for MBA
gb193092
 
Honest Reviews of Tim Han LMA Course Program.pptx
Honest Reviews of Tim Han LMA Course Program.pptxHonest Reviews of Tim Han LMA Course Program.pptx
Honest Reviews of Tim Han LMA Course Program.pptx
timhan337
 
The basics of sentences session 5pptx.pptx
The basics of sentences session 5pptx.pptxThe basics of sentences session 5pptx.pptx
The basics of sentences session 5pptx.pptx
heathfieldcps1
 
Francesca Gottschalk - How can education support child empowerment.pptx
Francesca Gottschalk - How can education support child empowerment.pptxFrancesca Gottschalk - How can education support child empowerment.pptx
Francesca Gottschalk - How can education support child empowerment.pptx
EduSkills OECD
 
How libraries can support authors with open access requirements for UKRI fund...
How libraries can support authors with open access requirements for UKRI fund...How libraries can support authors with open access requirements for UKRI fund...
How libraries can support authors with open access requirements for UKRI fund...
Jisc
 
TESDA TM1 REVIEWER FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...
TESDA TM1 REVIEWER  FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...TESDA TM1 REVIEWER  FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...
TESDA TM1 REVIEWER FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...
EugeneSaldivar
 
Model Attribute Check Company Auto Property
Model Attribute  Check Company Auto PropertyModel Attribute  Check Company Auto Property
Model Attribute Check Company Auto Property
Celine George
 
Guidance_and_Counselling.pdf B.Ed. 4th Semester
Guidance_and_Counselling.pdf B.Ed. 4th SemesterGuidance_and_Counselling.pdf B.Ed. 4th Semester
Guidance_and_Counselling.pdf B.Ed. 4th Semester
Atul Kumar Singh
 
Biological Screening of Herbal Drugs in detailed.
Biological Screening of Herbal Drugs in detailed.Biological Screening of Herbal Drugs in detailed.
Biological Screening of Herbal Drugs in detailed.
Ashokrao Mane college of Pharmacy Peth-Vadgaon
 
BÀI TẬP BỔ TRỢ TIẾNG ANH GLOBAL SUCCESS LỚP 3 - CẢ NĂM (CÓ FILE NGHE VÀ ĐÁP Á...
BÀI TẬP BỔ TRỢ TIẾNG ANH GLOBAL SUCCESS LỚP 3 - CẢ NĂM (CÓ FILE NGHE VÀ ĐÁP Á...BÀI TẬP BỔ TRỢ TIẾNG ANH GLOBAL SUCCESS LỚP 3 - CẢ NĂM (CÓ FILE NGHE VÀ ĐÁP Á...
BÀI TẬP BỔ TRỢ TIẾNG ANH GLOBAL SUCCESS LỚP 3 - CẢ NĂM (CÓ FILE NGHE VÀ ĐÁP Á...
Nguyen Thanh Tu Collection
 
Embracing GenAI - A Strategic Imperative
Embracing GenAI - A Strategic ImperativeEmbracing GenAI - A Strategic Imperative
Embracing GenAI - A Strategic Imperative
Peter Windle
 

Recently uploaded (20)

Thesis Statement for students diagnonsed withADHD.ppt
Thesis Statement for students diagnonsed withADHD.pptThesis Statement for students diagnonsed withADHD.ppt
Thesis Statement for students diagnonsed withADHD.ppt
 
The approach at University of Liverpool.pptx
The approach at University of Liverpool.pptxThe approach at University of Liverpool.pptx
The approach at University of Liverpool.pptx
 
Acetabularia Information For Class 9 .docx
Acetabularia Information For Class 9  .docxAcetabularia Information For Class 9  .docx
Acetabularia Information For Class 9 .docx
 
Introduction to AI for Nonprofits with Tapp Network
Introduction to AI for Nonprofits with Tapp NetworkIntroduction to AI for Nonprofits with Tapp Network
Introduction to AI for Nonprofits with Tapp Network
 
Unit 8 - Information and Communication Technology (Paper I).pdf
Unit 8 - Information and Communication Technology (Paper I).pdfUnit 8 - Information and Communication Technology (Paper I).pdf
Unit 8 - Information and Communication Technology (Paper I).pdf
 
special B.ed 2nd year old paper_20240531.pdf
special B.ed 2nd year old paper_20240531.pdfspecial B.ed 2nd year old paper_20240531.pdf
special B.ed 2nd year old paper_20240531.pdf
 
Azure Interview Questions and Answers PDF By ScholarHat
Azure Interview Questions and Answers PDF By ScholarHatAzure Interview Questions and Answers PDF By ScholarHat
Azure Interview Questions and Answers PDF By ScholarHat
 
Chapter 3 - Islamic Banking Products and Services.pptx
Chapter 3 - Islamic Banking Products and Services.pptxChapter 3 - Islamic Banking Products and Services.pptx
Chapter 3 - Islamic Banking Products and Services.pptx
 
Supporting (UKRI) OA monographs at Salford.pptx
Supporting (UKRI) OA monographs at Salford.pptxSupporting (UKRI) OA monographs at Salford.pptx
Supporting (UKRI) OA monographs at Salford.pptx
 
Marketing internship report file for MBA
Marketing internship report file for MBAMarketing internship report file for MBA
Marketing internship report file for MBA
 
Honest Reviews of Tim Han LMA Course Program.pptx
Honest Reviews of Tim Han LMA Course Program.pptxHonest Reviews of Tim Han LMA Course Program.pptx
Honest Reviews of Tim Han LMA Course Program.pptx
 
The basics of sentences session 5pptx.pptx
The basics of sentences session 5pptx.pptxThe basics of sentences session 5pptx.pptx
The basics of sentences session 5pptx.pptx
 
Francesca Gottschalk - How can education support child empowerment.pptx
Francesca Gottschalk - How can education support child empowerment.pptxFrancesca Gottschalk - How can education support child empowerment.pptx
Francesca Gottschalk - How can education support child empowerment.pptx
 
How libraries can support authors with open access requirements for UKRI fund...
How libraries can support authors with open access requirements for UKRI fund...How libraries can support authors with open access requirements for UKRI fund...
How libraries can support authors with open access requirements for UKRI fund...
 
TESDA TM1 REVIEWER FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...
TESDA TM1 REVIEWER  FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...TESDA TM1 REVIEWER  FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...
TESDA TM1 REVIEWER FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...
 
Model Attribute Check Company Auto Property
Model Attribute  Check Company Auto PropertyModel Attribute  Check Company Auto Property
Model Attribute Check Company Auto Property
 
Guidance_and_Counselling.pdf B.Ed. 4th Semester
Guidance_and_Counselling.pdf B.Ed. 4th SemesterGuidance_and_Counselling.pdf B.Ed. 4th Semester
Guidance_and_Counselling.pdf B.Ed. 4th Semester
 
Biological Screening of Herbal Drugs in detailed.
Biological Screening of Herbal Drugs in detailed.Biological Screening of Herbal Drugs in detailed.
Biological Screening of Herbal Drugs in detailed.
 
BÀI TẬP BỔ TRỢ TIẾNG ANH GLOBAL SUCCESS LỚP 3 - CẢ NĂM (CÓ FILE NGHE VÀ ĐÁP Á...
BÀI TẬP BỔ TRỢ TIẾNG ANH GLOBAL SUCCESS LỚP 3 - CẢ NĂM (CÓ FILE NGHE VÀ ĐÁP Á...BÀI TẬP BỔ TRỢ TIẾNG ANH GLOBAL SUCCESS LỚP 3 - CẢ NĂM (CÓ FILE NGHE VÀ ĐÁP Á...
BÀI TẬP BỔ TRỢ TIẾNG ANH GLOBAL SUCCESS LỚP 3 - CẢ NĂM (CÓ FILE NGHE VÀ ĐÁP Á...
 
Embracing GenAI - A Strategic Imperative
Embracing GenAI - A Strategic ImperativeEmbracing GenAI - A Strategic Imperative
Embracing GenAI - A Strategic Imperative
 

Learning Pulse - paper presentation at LAK17

  • 1. Learning Pulse D. Di Mitri, M. Scheffel, H. Drachsler, D. Börner, S. Ternier, M. Specht A machine learning approach for predicting performance in self-regulated learning using multimodal data Paper presentation at LAK17 15th March 2017, Vancouver, Canada
  • 2. Outline 1. Background, context, vision 2. Our approach 3. Data collection 4. Data analysis 5. Conclusions Pagina 2
  • 3. Data deluge in education Pagina 3
  • 4. Collecting learning experiences Picture from tincanapi.com Pagina 4
  • 6. Context: Self Regulated Learning Self-Regulated Learning → no guidance → no feedback → no support Pagina 6
  • 7. Vision: machine learning approach y = f(X) Learning Performance (output space) Predictive Model Multimodal Data (input space) Pagina 7
  • 9. Research questions (RQ-MAIN) How can we store, model and analyse multimodal data to predict performance in human learning? (RQ1) Which architecture allows the collection and storage of multimodal data in a scalable and efficient way? (RQ2) What is the best way to model multimodal data to apply supervise machine learning techniques? (RQ3) Which machine learning model is able to produce learner specific predictions on multimodal data? Pagina 9
  • 10. Participants • 9 PhD students at Welten institute • Different disciplines • Different working setups: – Time – Tasks – Operating systems Pagina 10
  • 11. Experimental timeline Pagina 11 Phase 0 Pre-test System architecture tested Phase 1 Training 3 weeks of data collection Phase 2 Validation 2 weeks of data collection and prediction
  • 12. Input space – multimodal data Pagina 12 Context Body Activities Body: physiological (heart-rate) and physical responses (steps) - from Fitbit HR Activities: applications used during learning from RescueTime Context: weather data from OpenWeatherMap
  • 13. Output space – Flow Csikszentmihalyi, 1972 Pagina 13 Theoretical Empirical
  • 14. Activity Rating Tool Productivity How productive was last activity? Stress How stressful was last activity? Challenge How challenging was last activity? Abilities How prepared did you feel for the activity? FLOW Participants rate hourly, from 7AM to 7PM A scalable web app! Client: Bootstrap + Jquery Sever: GoogleApp + Python “Very easy to use!” Pagina 14
  • 17. Berg, A., Scheffel, M., Drachsler, H., Ternier, S. & Specht, M. (2016). The Dutch xAPI Experience. Proceedings of the 6th International Conference on Learning Analytics and Knowledge (LAK’16), April 25-29, 2016, Edinburgh, UK. Data storing format for the Learning Record StoreExperience API
  • 20. Data collection • PULL data from the 3rd party APIs • Make the xAPI triples • PUSH data in the LRS • It’s scalable! • No collisions • It’s fast • It’s Interoperable Learning Pulse Server + Learning Record Store Pagina 20
  • 21. Data Processing Application Script in Python running on a VM which processes data in real time Pagina 21
  • 23. Transformed dataset • Time Series: tabular representation • 5 minutes intervals • Enough samples now! • Easier view for Machine Learning • Signal resampling needed 9410 observations X 29 attributes Pagina 23
  • 24. Issue 1) Feature extraction from Time Series Heart Rate Variability and Heart Rate Entropy… didn’t work SOLUTION • Mean of the signal • Maximum • Minimum • Standard Deviation • Average change Heart-ratesignalfor15mins Pagina 24
  • 25. Issue 2) Activity data very sparse Rule based grouping of applications Learners’ activity can be compared! Applications used are too sparse SOLUTION Let’s create application categories Pagina 25
  • 26. Issue 3) Number of labels available Trade-off: number of labels vs Seamlessness of the data collections NO SOLUTION Pagina 26
  • 27. Issue 5) Random vs continuous data Independence constraint Knowing one value of et for one observation does not help us to guess value of et+1 yt = α + βX t + et cov(et ,et+1) = 0 FIXED Effect RANDOM Effect SOLUTION follows... Pagina 27
  • 28. Mixed Effect Linear Model x0 x1 x2 ... xn-1 xn g y t0 x x x ... x x 1 y t1 x x x ... x x 1 y t2 x x x ... x x 2 y t... ... ... ... ... ... ... 2 y tp -1 x x x x x x 3 y tp ? ? --- --- --- --- x ? Random EffectsFixed Effects Group Used R-squared for goodness-test LIMITATIONS ● Convergence time ● Mono-output Pagina 28
  • 29. Issue 6) Inter-subject variability i.e. Participants have rated very differently SOLUTION Predictions are normalised wrt each learner xnew = (xmax – xmin) *xi/100 + xmin Pagina 29
  • 31. RQ1) Architecture The architecture developed was able of: 1. Importing great number of sensor data in xAPI format; 2. combining sensor data with self-reports 3. programmatically transform xAPI data 4. train predictive models & reuse them 5. save the predictions to compare with actual values Pagina 31
  • 32. RQ2) Represent multimodal data • Multiple Instance Representation • Each learning sample is a 5 minute interval • It’s suitable for machine learning Pagina 32
  • 33. RQ3) Machine learning model • Linear Mixed Effect Models allow 1. taking into account data specific to each learner 2. distinguish between fixed and random effects 3. Take categorical data into account. Pagina 33
  • 34. Limitations • Low accuracy of predictions R-Square tests Stress: 0.32, Challenge: 0.22, Flow score: 0.16, Abilities: 0.08, Productivity: 0.05. • Real-time issues Fitbit synchronisation, Virtual Machine performance • 3rd party API constraints • No great solution for grouping activity data (manual grouping) Pagina 34
  • 35. Opportunities • Data driven • Real Time feedback • Visualisations can show feedback • Seamless data collection • Multimodal dataset for reserach • Reusable architecture Pagina 35 *Börner, Tabuenca, Storm, Happe, and Specht. 2015 Example visualisation: The Feedback Cube*
  • 36. Q&A Thanks for listening! Daniele Di Mitri ddm@ou.nl @dimstudi0 Pagina 36 Check my poster!