Data analytics beyond data processing and how it affects Industry 4.0 Mathieu d’Aquin Vice Director, Insight Centre for Data Analytics, NUI Galway @mdaquin
Data analytics beyond data processing and how it affects Industry 4.0
Jan 21, 2018
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Data analytics beyond data processingand how it affects Industry 4.0Mathieu d’AquinVice Director, Insight Centre for Data Analytics, NUI Galway@mdaquin
Insight Centre for Data Analytics:- SFI-funded research centre- 5 department, in 4 universities- More than 40 partners- More than 400 researchers
Key areas of research:- Linked Data, Semantic Web
technologies, Web of Data, Open Data- Machine Learning & Statistics, Data
Mining, Knowledge Discovery- Media Analytics- Optimisation & Decision Analytics- Personal Sensing- Recommender Systems
Industry partnerships
Data at Web Scale
Interoperability, data integration
Automated reasoning, decision making from
streaming data
Data and Knowledge Infrastructure
Data Semantics
Industry 4.0?
1 2 3 4
Collaborative information platformsKnowledge extraction from text
Sentiment analysis, opinion miningFormalised knowledge sharing
knowledge discoveryautonomous knowledge agents
Modelisation, simulationData curation, cataloguing
Predictive analyticsMachine learning
Data miningStream processingEvent processing
Granular, adaptable privacy and securitySemantic interoperabilitySemantic interpretation
Example
Monitoring energy consumption from machines / appliances / devices.
Example
Monitoring energy consumption from machines / appliances / devices.
Stream
Processing
Event D
etection
Event P
rocessing
Decision S
upport
Example
Predicting future behaviour using machine learning.
Example
Detecting anomalous behaviour through derivation from the predicted behaviour.
Punctual anomaly (e.g. door
open)no alert
Back to normal (model and real values match)
Consistent anomaly (e.g. malfunction)
alert
Example
Data curation to
abstract anomalies
into common
templates....
dataset 1
dataset 2
dataset 3
…
dataset n
Example
Data curation to
abstract anomalies
into common
templates....
dataset 1
dataset 2
dataset 3
…
dataset n
Example
Data curation to
abstract anomalies
into common
templates....
dataset 1
dataset 2
dataset 3
…
dataset n
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
...
Example
Data curation to
abstract anomalies
into common
templates....
dataset 1
dataset 2
dataset 3
…
dataset n
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
...Data mining
(formal concept analysis)
Anomalous Data
Templates
Example
...
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
...
Data mining (formal concept
analysis)
Anomalous Data
Templates
Example
...
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
...
Data mining (formal concept
analysis)
Anomalous Data
Templates
Anomaly signatures
Example
...
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
...
Data mining (formal concept
analysis)
Anomalous Data
Templates
Anomaly signatures
Engineering & maintenance
reports (text)
Example
...
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
...
Data mining (formal concept
analysis)
Anomalous Data
Templates
Anomaly signatures
Engineering & maintenance
reports (text)
Semantic classificat
ion
Anomaly typologyontology
Example
...
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
Data annotation, metadata
...
Data mining (formal concept
analysis)
Anomalous Data
Templates
Anomaly signatures
Engineering & maintenance
reports (text)
Semantic classificat
ion
Anomaly typologyontology
Decision(predictive
maintenance)
Example
Uses stream processing, event processing, decision support, machine learning, data mining, knowledge discovery, data curation, data cataloguing, semantic annotation, reasoning, data profiling, text analytics…
… from various projects in various domains.
Conclusion
Data analytics and the meaningful, semantic interpretation of data from diverse sources is critical to manufacturing and all other branches of engineering.
Insight(@NUIG) research and develop the key technologies to make smart approaches exploitable by industry, from large scale tech. companies, to innovative SMEs.
insight-centre.org
nuig.insight-centre.org
mdaquin.net
@mdaquin
we are always excited to talk to potential partners, and to new talents.
Related Documents
