Download Workshop: Introduction to Educational Data Mining

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Nonlinear dimensionality reduction wikipedia , lookup

Transcript
Workshop: Introduction to
Educational Data Mining
Yiling Zeng & Shirley Alexander
Educational Data Mining (EDM) is an emerging area of research concerned with
the use of data mining techniques in education in order to predict student
performance, provide recommendations on optimal student interventions, and
develop models of student behaviour. It is one of a number of contemporary
approaches to learning analytics which relies on the development of data mining
techniques.
This workshop will commence with a brief overview of the evolution of the
Intelligent Tutoring Systems of the 1980s which attempted to build student
models to today’s Educational Data Mining techniques. A variety of techniques
on the prediction of student performance, student modelling, student grouping,
social network analysis and feedback and recommendation providing will also
be introduced in this workshop.
A case study on a recently-developed educational data mining system, i-Educator,
which is able to predict students’ academic risk levels with potential reasons will
be used to illustrate the use of various EDM techniques. The analysis of iEducator on particular subjects with high failure rates will be used as examples
to demonstrate the use of the following EDM techniques:
 Contrast pattern mining;
 Voting forest (A forest of decision trees)
 Ensemble Methods for integrating different models
Workshop outcomes: participants will be able to



Articulate the evolution of techniques involved in the modelling of
student behaviour student
Compare a range of basic educational data mining techniques
Develop a plan to model student behaviour
Workshop approach:



Presentation and discussion of Intelligent Tutoring Systems – successes
and failures
Case Study – An analysis of an educational data mining system which
models students’ behaviours on ”killer subjects” to demonstrate the
relevant EDM techniques
Interactions with exercises based on the concepts and algorithms
introduced.
Relevant references and resources:





Cristobal Romero and Sebastian Ventura: Educational Data Mining: A
Review of the State-of-the-Art. Transactions on Systems, Man, and
Cybernetics – Part C: Applications and Reviews. 40(6) pp: 601 – 618,
2010.
Ryan S.J.D. Baker and Kalina Yacef: The State of Educational Data Mining
in 2009: A Review and Future Visions. Journal of Educational Data Mining,
1 (1), pp:3-17, 2009.
International Conference on Educational Data Mining.
http://www.educationaldatamining.org/proceedings.html
Journal of Educational Data Mining.
http://www.educationaldatamining.org/JEDM/index.php
IEEE Task Force on Educational Data Mining.
http://datamining.it.uts.edu.au/edd/
About the facilitators
Yiling Zeng is a postdoc research associate at Advanced Analytics Institute, UTS.
His research includes educational data mining, machine learning and text mining.
He has worked on several student analytics projects, including the UTS Vice
Chancellor’s Learning and Teaching Grant Phase I and Phase II.
Shirley Alexander is Deputy Vice-Chancellor (Teaching,Learning & Equity) at
UTS. She has undergraduate and postgraduate qualifications in Statistics and
Artificial Intelligence and is leading the university-wide project for UTS to
become a “data intensive university”.