Recording
The workshop has finished, you can find a video of it at http://youtu.be/ncQP8YsYseY or below.

Introduction
The goal of this tutorial is to share data mining tools and techniques used by computer scientists with educational social scientists. We broadly define educational social scientists as being made up of people with backgrounds in the learning sciences, cognitive psychology, and educational research. The learning analytics community is heavily populated with researchers of these backgrounds, and we believe those that find themselves at the intersection of research, theory, and practice have a particular interest in expanding their knowledge of data-driven tools and techniques.

Objectives
By the end of this course we expect that attendees are able to: 
  1. Describe the differences between supervised and unsupervised classification 
  2. Understand how to choose a classification method for a particular research question 
  3. Frame different kinds of educational datasets in a way that is appropriate for data mining 
  4. Contextualize the results of a J48 decision tree to their research questions 
  5. Contextualize the results of k-means clustering to their research questions 
  6. Apply knowledge of the Weka toolkit to create decision trees or clusters of new educational datasets
Setup
Schedule
This half-day tutorial is schedule for the afternoon of Monday June 30th.  Registration free with admittance to LASI 2014A preliminary schedule for this half-day tutorial is as follows: 

1:15 - 1:30
Casual welcome, come and download dataset.
1:30 - 2:00
A short overview of what kinds of educational problems classification in data mining has been applied to, along with a description of research questions that classification is not well suited to answering.
2:00 - 2:30
An introduction to what cluster analysis is, and a guided example of applying k-means to real datasets. Attendees will be walked through the process, and will cluster the data using various kinds of parameters.
2:30 - 3:00
A discussion of understanding the results of clustering. E.g. what is a centroid, how to measure the quality of the cluster analysis, applications of clustering in other educational areas.
3:00 - 3:30
Coffee Break
3:30 - 4:10
A guided example of building J48 decision trees using Weka. Attendees will be given access to real datasets, and walked through the various parameters (leaf size, ten fold testing, etc.) available in the Weka toolkit.
4:10 - 4:30
A discussion of understanding the results of the decision tree process, such as how to read the confusion matrix, understanding rules in the tree, etc. Participants will be encourage to explore parameters of the classification, and relate these to outcomes.
4:30 - 5:00
A wrap up discussion reviewing the objectives of the tutorial, next steps for participants who wish to learn more, and an introduction to exciting new open MOOC datasets that participants can explore in more detail.

Presenters
Dr. Christopher Brooks is a Research Fellow at the University of Michigan School of Information with an interest in learning analytics and quantifying the effects of educational technology on the teaching and learning process. He earned his PhD in Computer Science from the University of Saskatchewan in 2012. 

Dr. Zachary Pardos is an Assistant Professor at UC Berkeley in a joint position between the School of Information and Graduate School of Education. He earned his PhD in Computer Science at Worcester Polytechnic Institute in the Tutor Research Group in 2012.

Vitomir Kovanovic is a PhD candidate and research assistant in the School of Interactive Arts and Technology at Simon Fraser University. He received his M.Sc degree in Computer Science and Software Engineering in 2011 and his B.Sc degree in Information Systems and Business Administration in 2009 from the University of Belgrade, Serbia.  His research interests include learning analytics, educational data mining, self-regulated learning and online education. He is a member of the Society for Learning Analytics Research and a member of program committees of several conferences in technology-enhanced learning. In his PhD research, he focuses on the use of learning trace data for understanding the effects of technology use on the quality of social learning process and learning outcomes.

Srecko Joksimovic is a PhD student and research assistant at School of Interactive Arts and Technology, Simon Fraser University. He is
a member of Society for Learning Analytics and actively involved in Semantic Web technologies research. He received B.Sc and M.Sc in Information Systems at Military Academy, Belgrade, Serbia. His research interests are in the areas of machine learning, natural language processing, ducational data mining, and learning analytics. In his PhD research, he focuses on analyzing to what extent learner generated data can be used to predict learning outcome.