Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Recommender Systems and Learning Analytics

Presentation to support the Seminar Talk at RWTH Aachen University
by

Omar Sanchez

on 3 July 2011

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Recommender Systems and Learning Analytics

Recommender Systems and
Learning Analytics Vlatko L. and W. Omar S. Agenda Motivation
Recommender Systems
Educational Data Mining & Learning Analytics
Synergies and Comparison
Future work
Conclusion New ways to use existing systems and information Motivation Current education trends are changing Research on e-Learning and learner interaction with the system is a relatively new topic Production of large amounts of data, as byproduct of system interaction Recommender Systems What is a Recommender System? Content based Collaborative filtering Hybrid Educational Data Mining & Learning Analytics Motivation Background Objectives Different target information Emerging from conventional data mining Taxonomy Task Domain Implementations Concerns big brother
holistic view
faculty involvement
profiling Definitions Business Analytics
Web Analytics
Educational Data Mining
Academic Analytics
Action Analysis Stakeholders Students, Learners, Pupils
Faculty, Professors, Teachers, Educators
Curriculum Designers, Course Developers, Researchers
Executive Officers, School Administrators Synergies and Comparison Decouple Recommender Systems, Educational Data Mining and Learning Analytics Comparison Different dataset
Different set of possible recommendations

No longer responsible for processing raw data
Consume data from EDM and LA system A new approach towards developing learning platforms Future Work Educational Data Mining Learning Analytics Development of standard data formats

EDM tools for non-experts

Integration with e-Learning systems

Adapt traditional mining algorithms to the educational domain Resolve privacy and ownership issues

Understand the potential and power of the collected data

Real time feedback

Bridge gap between information technology staff and non-technical staff General EDM-LA-Recommender System connection

Define Recommender Systems' task

Clearly define goals and measure the success of the Recommender Systems

Integration
User acceptance Conclusions Agenda Introduction
Recommender Systems
Educational Data Mining & Learning Analytics
Synergies and Comparison
Future work
Conclusion Agenda Motivation
Recommender Systems
Educational Data Mining & Learning Analytics
Synergies and Comparisons
Future work
Conclusion Agenda Motivation
Recommender Systems
Educational Data Mining & Learning Analytics
Synergies and Comparison
Future work
Conclusion Agenda Motivation
Recommender Systems
Educational Data Mining & Learning Analytics
Synergies and Comparison
Future work
Conclusion Agenda Motivation
Recommender Systems
Educational Data Mining & Learning Analytics
Synergies and Comparison
Future work
Conclusion Long term research is needed

Wider acceptance is required for success

Boundaries need to be determined Agenda Motivation
Recommender Systems
Educational Data Mining & Learning Analytics
Synergies and Comparison
Future work
Conclusion Similarities and Differences - A Short Overview Existing evaluation approaches are not sufficient Emerging use of IT solutions to enhance education Student-teacher interaction is changing Improving learning processes Improve sucess rates amongst certain groups of students EDM LA Traditional Data Mining approaches Combination of already existing analytical approaches to reach specific educational goals EDM LA gathers data from different learning platforms, separates the useful data, structures it and provides meaningful interpretation and information data clustering and analysis using modified mining techniques to suit the educational environment Converting raw data into useful information EDM LA Predict student success Determine actions based on success Use data to shape and improve educational performance Improve teaching and mentoring i.e. Provide an empirical base for decision making Mine institutional data to produce actionable intelligence But: Shift towards the main objectives of Learning Analytics Assess students' performance Study the pedagogical support of learning software systems Data analysis to provide feedback to teachers/instructors LA EDM data privacy
data stewardship
information sharing
obligation to act
distribution of resources Build a good Learning Analytics System

Current systems are data rich, but information poor Propose guidelines and design patterns Web Based Learning systems

Address the most immediate stakeholders

Students
Faculty Define and design methods for extracting the needed information Resolve issues with synchronization methods between methods and different sets of data Extensibility on the fly Build/Modify good Recommender System Seminar Recommender Systems - Prof. Dr. Ulrik Schroeder SS 2011 Student Affairs
Organizations, Universities, Learning Providers
System Administrators, Network Administrators, Information Technology Officers
Full transcript