Schedule
| Start Time | End Time | Subject |
| 9:00am | 9:30am | LearnSphere Overview High-level overview of the tools and features that LearnSphere provides researchers. |
| 9:30am | 10:30am | Importing Data (Lecture/Demo & Hands-On) We will discuss and walk through the process of importing new data into DataShop. We will discuss custom fields that allow researchers to include additional data that is not currently represented in the current data model. Attendees will then have the opportunity to import their own datasets and receive hands-on assistance as needed. |
| 10:30am | 11:00am | COFFEE BREAK |
| 11:00am | 12:30pm | Learning Curve Analysis (Lecture/Demo & Hands-On) Learning curve visualizations and parameter estimates are workflow outputs that researchers can use to answer research questions and derive modeling insights. We will demonstrate how to use the LearnSphere learning curve tools and visualizations to analyze workflow outputs. |
| 12:30pm | 1:30pm | LUNCH |
| 1:30pm | 2:00pm | The Distributed Datashop We will present developments towards a distributed Datashop, with focus on the first new branch hosted at the University of Memphis. |
| 2:00pm | 2:30pm | Workflow Tools Overview |
| 2:30pm | 3:00pm | Building Parallel Workflows with AFM and BKT (Demo) |
| 3:00pm | 3:30pm | COFFEE BREAK |
| 3:30pm | 4:00pm | Building Parallel Workflows (Hands-On) |
| 4:00pm | 4:30pm | Creating a Custom Workflow (Lecture/Demo) |
| 4:30pm | 5:00pm | Discussion & Feedback on LearnSphere Open discussion about attendees’ hands-on experiences applying the various workflow and learning curve tools in LearnSphere. We will solicit feedback and suggestions for improvements. |
Overview
This tutorial will explore the application of novel educational data mining workflows using LearnSphere, a new $5 million NSF funded data sharing and analysis portal that extends the existing DataShop infrastructure and includes teams from Carnegie Mellon, Stanford, Memphis, and MIT. Increased flexibility to accommodate custom workflows is one of the core ways in which LearnSphere expands upon DataShop. LearnSphere’s goal is to support any custom analysis workflow that can be applied to the educational datasets in DataShop and to produce standardized workflow outputs that facilitate quantitative and qualitative model comparisons. As a result, researchers will be able to apply their own analysis workflows to the vast array of datasets available in the DataShop repository. It affords researchers the advantages of (1) using the DataShop learning curve visualizations and cognitive model improvement tools on the outputs of their own models, (2) easily comparing their models both quantitatively and graphically to any alternative models that are currently in LearnSphere (e.g., AFM, BKT, PFA, and a growing library of variants) or any other models that have been uploaded to LearnSphere as a custom workflow, and (3) sharing their models with the research community. Researchers will be able to compare, across alternative models, not only model fit metrics like AIC, BIC, and cross validation but also the parameter estimates themselves. In this way, LearnSphere can potentially benefit the broader EDM community by promoting better standardization and management over data mining workflows.
This tutorial will provide an overview of the LearnSphere architecture, with a focus on the custom workflow tools. Participants will learn how to run different analysis workflows, including their own custom analyses/models if desired, and leverage LearnSphere's learning curve visualization and model parameter report tools. They will also learn how to locate interesting existing datasets in the LearnSphere data repository and how to import their own data for analyses. Funding is available to cover conference and tutorial registration fees for some attendees.
Organizers
Ran Liu, Carnegie Mellon University
Philip Pavlik, University of Memphis
John Stamper, Carnegie Mellon University
Michael Eagle, Carnegie Mellon University
Questions?
Contact John Stamper—john AT stamper DOT org
Equipment needed
Participants should bring laptop computers.