Using Linked Data in Learning Analytics
1 like703 views
This document provides an overview of a tutorial on using linked data in learning analytics. The tutorial aims to teach researchers and developers the basics of exploiting linked data resources to enrich learning analytics processes. It introduces linked data technologies like RDF and SPARQL and includes hands-on exercises using real education datasets. The tutorial also explores how tools like R, Tableau and Gephi can interface with linked data. It is supported by the LinkedUp project, which provides scenario data and a framework for evaluating linked data applications in education.
![Using Linked Data in Learning Analytics
Mathieu d’Aquin
Stefan Dietze
Knowledge Media Institute
The Open University, UK
L3S Research Center
Leibniz University, Hanover,
Germany
mathieu.daquin@open.ac.uk
Hendrik Drachsler
dietze@l3s.de
Eelco Herder
CELSTEC
Open University of the
Netherlands
L3S Research Center
Leibniz University, Hanover,
Germany
hendrik.drachsler@ou.nl
ABSTRACT
“Using Linked Data in Learning Analytics” is a tutorial targeting researchers in Learning Analytics interested in exploiting linked data resources, developers of Learning Analytics solutions that could benefit from Linked Data and
data owners wanting to understand how linked data can help
the analysis of their data in relation to other sources of information. The tutorial is described in more details at http://
linkedu.eu/event/lak2013-linkeddata-tutorial/, where
learning material related to the topic of the tutorial will also
be disseminated.
Categories and Subject Descriptors
H.3.1 [Information Storage and Retrieval]: Content
Analysis and Indexing
General Terms
Design, Experimentation
Keywords
leaning analytics, data mining, linked data, visualisation,
tutorial
1.
WHY LINKED DATA
Linked Data [5] is a set of principles and technologies
aimed at using the architecture of the web to share, expose
and integrate data in a global, collaborative space. The
success of the idea is undeniable, with a large variety of
organisations (including governments, broadcasters, museums, libraries and, of course, universities [2]) now exposing
their data through Linked Data technologies online for everybody to use. This massive movement towards the Web of
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. To copy otherwise, to
republish, to post on servers or to redistribute to lists, requires prior specific
permission and/or a fee.
Copyright 20XX ACM X-XXXXX-XX-X/XX/XX ...$15.00.
herder@l3s.de
Data naturally impacts on Learning Analytics activities (see
the LAK 2012 workshop on Leaning Analytics and Linked
Data [4]), by providing not only new, more flexible ways of
integrating and manipulating data, standardising datasets,
but also a vast sources of data that can enrich to analysis
and mining activities for example with information regarding the learner’s context (location, time, related events, institutions, etc.) and the resources related to the learning
experience (open educational resources, books and base material, cultural resources, etc. – see for example [3])
2.
AIM OF THE TUTORIAL
This tutorial provides Learning Analytics practitioners
with the basic knowledge and skills required to exploit the
new possibilities offered by linked data, especially through
exploiting the wealth of data sources already available in the
linked data cloud. It introduces the basic technologies and
practices generally associated with Linked Data, including
graph-based data modelling with RDF1 and relevant vocabularies (see [1]), data discovery on the linked data cloud and
the use of linked data endpoints (with SPARQL2 ). As the focus of the tutorial is on the concrete use of these technologies
and practices within a Learning Analytics scenario, a large
part of the sessions are dedicated to hands-on exercises with
data and use cases of relevance to Learning Analytics (from
LinkedUniversities.org, LinkedEducation.org, and the
LinkedUp support action3 ).
In addition to addressing the basic skill-set a Learning
Analytics practitioner might require in their use of Linked
Data, the tutorial focuses on practical ways in which Linked
Data resources can be exploited in Learning Analytics processes, especially through common tools interfacing with
Linked Data technologies. This is achieved through handson tasks where these tools are used to analyse data originating and/or integrated with the Linked Data cloud. Such
tasks range from the manipulation of data in common tools
(e.g. spreadsheets), to employing more advanced statistical
or network analysis methods with R4 , Tableau5 and Gephi6 ,
1
http://www.w3.org/RDF/
http://www.w3.org/TR/rdf-sparql-query/
3
http://linkedup-project.eu
4
http://www.r-project.org/
5
http://www.tableausoftware.com/
6
https://gephi.org/
2](https://image.slidesharecdn.com/using-linked-data-in-learning-analytics4778-131114053748-phpapp01/85/Using-Linked-Data-in-Learning-Analytics-1-320.jpg)
![• Basics of manipulating linked data: Hands-on
tasks guided by the given scenario (first shortly presented).
through interfaces to import Linked Data.
3.
CONNECTION WITH LINKEDUP
• Using linked data in analytics tools: Presentation
of available toolset and hands-on tasks on using these
tools in the given scenario, with the given data as well
as other data identified in the previous parts.
The tutorial is supported by the LinkedUp support action,
which provides illustrative scenarios and the corresponding
datasets, with the benefit of being concretely “endorsed”
by members of the network of organisations associated to
LinkedUp. Indeed, one of the key product of LinkedUp is a
data repository providing a structured and organised collection of datasets of relevance for education, that are properly
catalogued and mapped.
In addition, LinkedUp is organising a large scale competition for applications of web data-related approaches for education. The tutorial is an occasion to investigate the base
techniques making the development of such applications possible, as well as a way to discuss and brainstorm ideas for
novel linked data-based educational services. Concretely,
this connection is materialised through the inclusion in the
tutorial of results and discussions related the the “LAK Data
Challenge”7 .
4.
• Getting Data for Education with LinkedUp: More
detailed tasks focusing on the use of the LinkedUp data
pool to address the data needs of the use case scenarios
identified in the previous parts.
• Evaluating Linked Data Applications with
LinkedUp: Tasks on using the LinkedUp Evaluation framework for assessing applications of the Web
of Data in educational scenarios.
• The LinkedUp - LAK Corpus Challenge: Moving
towards more concrete examples of use of linked open
data for education and analytics, this part includes
presentations and demonstrations from the LAK Corpus Challenge.
THE LAK DATA CHALLENGE
The LAK Dataset8 provides access to structured metadata from research publications in the field of learning analytics. Beyond merely publishing the data, the LAK Data
Challenge is meant to actively encourage the use of such
data in innovative applications. The LAK Data Challenge is
sponsored by the LinkedUp support action and is co-located
with the ACM LAK13 Conference9 .
The main objective of the challenge is to understand what
analytics on learning analytics can tell us: How can we make
sense of this emerging field’s historical roots, current state,
and future trends, based on how its members report and
debate their research This may include submissions which
cover the analysis and assessment of the emerging LAK community in terms of topics, people, citations or connections
with other fields; Innovative applications to explore, navigate and visualise the dataset (and/or its correlation with
other datasets); Usage of the dataset as part of recommender
systems, etc.
• Going Further: Presentation of existing initiatives
such as LinkedUniversities.org and LinkedEducation.
org, the LinkedUp challenge, summer schools and courses
on Linked Data, etc. that can help moving beyond the
content of the proposal in the exploitation of linked
data in leaning analytics scenarios.
Acknowledgement
This tutorial is supported by the LinkedUp support action –
Linking Web Data for Education Project; SOLAR10 – Society for Learning Analytics Research; and EATEL SIG dataTEL11 .
6.
REFERENCES
[1] M. d’Aquin. Linked Data for Open and Distance
Learning. Commonwealth of Learning report –
Available from http://www.col.org/resources/
publications/Pages/detail.aspx?PID=420, 2012.
[2] M. d’Aquin. Putting linked data to use in a large
higher-education organisation. In Proceedings of the
5. OVERVIEW OF THE TUTORIAL
Interacting with Linked Data (ILD) workshop at
The tutorial is a mix of presentations, interactive sessions
Extended Semantic Web Conference (ESWC), 2012.
and hands-on activity. Below is a rough outline:
[3] M. d’Aquin and N. Jay. Interpreting data mining results
with linked data for learning analytics: Motivation,
• Introduction to Linked Data and relevance to
case study and direction. In International Learning
Learning Analytics: A brief presentation of the baAnalytics and Knowledge Conference, LAK, 2013.
sic notions of linked data and on the way it can be
[4] H. Drachsler, S. Dietze, W. Greller, M. d’Aquin,
used within a Learning Analytics process.
Jovanovic, A. J., Pardo, W. Reinhardt, and K. Verbert.
1st international workshop on learning analytics and
• Linked Data for Learning Analytics Scenario:
linked data. In S. Dawson, C. Haythornthwaite, S. B.
Together with the participants, list ways in which data
Shum, D. Gasevic, and R. Fergusson, editors, 2nd
available on the linked data cloud could be integrated/useful
International Conference on Learning Analytics and
to a Learning Analytics process. Inspecting the Linked
Knowledge, LAK, 2012.
Data cloud, identify the types of data that could be
[5] T. Heath and C. Bizer. Linked Data: Evolving the Web
useful in such scenarios.
into a Global Data Space. Synthesis Lectures on the
7
http://www.solaresearch.org/events/lak/
Semantic Web: Theory and Technology. Morgan &
lak-data-challenge/
Claypool, 2011.
8
http://www.solaresearch.org/resources/
10
lak-dataset/
http://www.solaresearch.org
9
11
http://lakconference2013.wordpress.com/
http://www.ea-tel.eu/special-interest-groups/](https://image.slidesharecdn.com/using-linked-data-in-learning-analytics4778-131114053748-phpapp01/85/Using-Linked-Data-in-Learning-Analytics-2-320.jpg)
Ad
More Related Content
What's hot (20)
Viewers also liked (6)
Ad
Similar to Using Linked Data in Learning Analytics (20)
Ad
More from Hendrik Drachsler (20)
Using Linked Data in Learning Analytics
- 1. Using Linked Data in Learning Analytics Mathieu d’Aquin Stefan Dietze Knowledge Media Institute The Open University, UK L3S Research Center Leibniz University, Hanover, Germany [email protected] Hendrik Drachsler [email protected] Eelco Herder CELSTEC Open University of the Netherlands L3S Research Center Leibniz University, Hanover, Germany [email protected] ABSTRACT “Using Linked Data in Learning Analytics” is a tutorial targeting researchers in Learning Analytics interested in exploiting linked data resources, developers of Learning Analytics solutions that could benefit from Linked Data and data owners wanting to understand how linked data can help the analysis of their data in relation to other sources of information. The tutorial is described in more details at http:// linkedu.eu/event/lak2013-linkeddata-tutorial/, where learning material related to the topic of the tutorial will also be disseminated. Categories and Subject Descriptors H.3.1 [Information Storage and Retrieval]: Content Analysis and Indexing General Terms Design, Experimentation Keywords leaning analytics, data mining, linked data, visualisation, tutorial 1. WHY LINKED DATA Linked Data [5] is a set of principles and technologies aimed at using the architecture of the web to share, expose and integrate data in a global, collaborative space. The success of the idea is undeniable, with a large variety of organisations (including governments, broadcasters, museums, libraries and, of course, universities [2]) now exposing their data through Linked Data technologies online for everybody to use. This massive movement towards the Web of Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Copyright 20XX ACM X-XXXXX-XX-X/XX/XX ...$15.00. [email protected] Data naturally impacts on Learning Analytics activities (see the LAK 2012 workshop on Leaning Analytics and Linked Data [4]), by providing not only new, more flexible ways of integrating and manipulating data, standardising datasets, but also a vast sources of data that can enrich to analysis and mining activities for example with information regarding the learner’s context (location, time, related events, institutions, etc.) and the resources related to the learning experience (open educational resources, books and base material, cultural resources, etc. – see for example [3]) 2. AIM OF THE TUTORIAL This tutorial provides Learning Analytics practitioners with the basic knowledge and skills required to exploit the new possibilities offered by linked data, especially through exploiting the wealth of data sources already available in the linked data cloud. It introduces the basic technologies and practices generally associated with Linked Data, including graph-based data modelling with RDF1 and relevant vocabularies (see [1]), data discovery on the linked data cloud and the use of linked data endpoints (with SPARQL2 ). As the focus of the tutorial is on the concrete use of these technologies and practices within a Learning Analytics scenario, a large part of the sessions are dedicated to hands-on exercises with data and use cases of relevance to Learning Analytics (from LinkedUniversities.org, LinkedEducation.org, and the LinkedUp support action3 ). In addition to addressing the basic skill-set a Learning Analytics practitioner might require in their use of Linked Data, the tutorial focuses on practical ways in which Linked Data resources can be exploited in Learning Analytics processes, especially through common tools interfacing with Linked Data technologies. This is achieved through handson tasks where these tools are used to analyse data originating and/or integrated with the Linked Data cloud. Such tasks range from the manipulation of data in common tools (e.g. spreadsheets), to employing more advanced statistical or network analysis methods with R4 , Tableau5 and Gephi6 , 1 http://www.w3.org/RDF/ http://www.w3.org/TR/rdf-sparql-query/ 3 http://linkedup-project.eu 4 http://www.r-project.org/ 5 http://www.tableausoftware.com/ 6 https://gephi.org/ 2
- 2. • Basics of manipulating linked data: Hands-on tasks guided by the given scenario (first shortly presented). through interfaces to import Linked Data. 3. CONNECTION WITH LINKEDUP • Using linked data in analytics tools: Presentation of available toolset and hands-on tasks on using these tools in the given scenario, with the given data as well as other data identified in the previous parts. The tutorial is supported by the LinkedUp support action, which provides illustrative scenarios and the corresponding datasets, with the benefit of being concretely “endorsed” by members of the network of organisations associated to LinkedUp. Indeed, one of the key product of LinkedUp is a data repository providing a structured and organised collection of datasets of relevance for education, that are properly catalogued and mapped. In addition, LinkedUp is organising a large scale competition for applications of web data-related approaches for education. The tutorial is an occasion to investigate the base techniques making the development of such applications possible, as well as a way to discuss and brainstorm ideas for novel linked data-based educational services. Concretely, this connection is materialised through the inclusion in the tutorial of results and discussions related the the “LAK Data Challenge”7 . 4. • Getting Data for Education with LinkedUp: More detailed tasks focusing on the use of the LinkedUp data pool to address the data needs of the use case scenarios identified in the previous parts. • Evaluating Linked Data Applications with LinkedUp: Tasks on using the LinkedUp Evaluation framework for assessing applications of the Web of Data in educational scenarios. • The LinkedUp - LAK Corpus Challenge: Moving towards more concrete examples of use of linked open data for education and analytics, this part includes presentations and demonstrations from the LAK Corpus Challenge. THE LAK DATA CHALLENGE The LAK Dataset8 provides access to structured metadata from research publications in the field of learning analytics. Beyond merely publishing the data, the LAK Data Challenge is meant to actively encourage the use of such data in innovative applications. The LAK Data Challenge is sponsored by the LinkedUp support action and is co-located with the ACM LAK13 Conference9 . The main objective of the challenge is to understand what analytics on learning analytics can tell us: How can we make sense of this emerging field’s historical roots, current state, and future trends, based on how its members report and debate their research This may include submissions which cover the analysis and assessment of the emerging LAK community in terms of topics, people, citations or connections with other fields; Innovative applications to explore, navigate and visualise the dataset (and/or its correlation with other datasets); Usage of the dataset as part of recommender systems, etc. • Going Further: Presentation of existing initiatives such as LinkedUniversities.org and LinkedEducation. org, the LinkedUp challenge, summer schools and courses on Linked Data, etc. that can help moving beyond the content of the proposal in the exploitation of linked data in leaning analytics scenarios. Acknowledgement This tutorial is supported by the LinkedUp support action – Linking Web Data for Education Project; SOLAR10 – Society for Learning Analytics Research; and EATEL SIG dataTEL11 . 6. REFERENCES [1] M. d’Aquin. Linked Data for Open and Distance Learning. Commonwealth of Learning report – Available from http://www.col.org/resources/ publications/Pages/detail.aspx?PID=420, 2012. [2] M. d’Aquin. Putting linked data to use in a large higher-education organisation. In Proceedings of the 5. OVERVIEW OF THE TUTORIAL Interacting with Linked Data (ILD) workshop at The tutorial is a mix of presentations, interactive sessions Extended Semantic Web Conference (ESWC), 2012. and hands-on activity. Below is a rough outline: [3] M. d’Aquin and N. Jay. Interpreting data mining results with linked data for learning analytics: Motivation, • Introduction to Linked Data and relevance to case study and direction. In International Learning Learning Analytics: A brief presentation of the baAnalytics and Knowledge Conference, LAK, 2013. sic notions of linked data and on the way it can be [4] H. Drachsler, S. Dietze, W. Greller, M. d’Aquin, used within a Learning Analytics process. Jovanovic, A. J., Pardo, W. Reinhardt, and K. Verbert. 1st international workshop on learning analytics and • Linked Data for Learning Analytics Scenario: linked data. In S. Dawson, C. Haythornthwaite, S. B. Together with the participants, list ways in which data Shum, D. Gasevic, and R. Fergusson, editors, 2nd available on the linked data cloud could be integrated/useful International Conference on Learning Analytics and to a Learning Analytics process. Inspecting the Linked Knowledge, LAK, 2012. Data cloud, identify the types of data that could be [5] T. Heath and C. Bizer. Linked Data: Evolving the Web useful in such scenarios. into a Global Data Space. Synthesis Lectures on the 7 http://www.solaresearch.org/events/lak/ Semantic Web: Theory and Technology. Morgan & lak-data-challenge/ Claypool, 2011. 8 http://www.solaresearch.org/resources/ 10 lak-dataset/ http://www.solaresearch.org 9 11 http://lakconference2013.wordpress.com/ http://www.ea-tel.eu/special-interest-groups/