Alongside 2021 Educational Data Mining Conference in Paris, France

A Workshop on Process Analysis Methods

For Educational Data

29 June 2021, Virtual, Register here

Papers and code

Link to the papers: https://drive.google.com/drive/u/0/folders/1O_zuCxVEPY3zroUMpur53YmDtw57mCG7

seqClustR: https://github.com/aditya9352/seqClustR
fuzzymineR:
https://github.com/nirmalpatel/fuzzymineR
(please reach out to nirmal@playpowerlabs.com you need anything else)

Agenda

(The times indicate how much time a particular agenda item will last for)

  • 5 min - Greetings and workshop overview

  • 25 min - Introduction to Educational Process Analysis

  • 40 min - Panel discussion: How can we use process data to do innovative learning science and technology research?

  • 25 min - Understanding Students’ Problem-Solving Processes via Action Sequence Analyses

  • 25 min - Using Markov Transition Matrix to Analyze Parsons Puzzle Solutions

  • 30 min - Break

  • 25 min - LOGANShiny: An app for illustrating process data analysis from international large-scale assessments

  • 25 min - Exploratory Process Analysis of Teacher Learning of AI Integration through Collaborative Design

  • 25 min - seqClustR: An R Package for Sequence Clustering

  • 40 min - Open Discussion, topics that are common across papers and are general to process analysis

  • 5 min - Concluding notes

Schedule

29 June 2021

US EST: 10:00 AM to 2:30 PM
US PST: 7:00 AM to
11:30 AM
Central European Time: 4:00 PM to
8:30 PM
Indian Standard Time: 7:30 PM to 12:
00 AM

Introduction


We are happy to announce a Process Analysis in Education workshop at the 2021 Educational Data Mining Conference in Paris, France. We welcome all researchers who are interested in discovering the learning and teaching processes from educational data in order to advance our understanding of human learning and build more effective educational programs and technology.


This workshop aims to bring together researchers using different methods to analyze process aspects of educational data. The ubiquity of temporal student data has enabled us to apply many different process analysis methods to uncover educational processes hidden within the data. The choice of the process analysis method often depends on the research questions and the form of the data, and they vary widely among researchers. Through this workshop, we are creating an event where researchers can share the process analysis methods they use, and discover what other methods are being used in the community. We believe that this sharing of knowledge will empower our community to discover novel insights from educational data.


(Read the accepted workshop proposal)

A graph showing how students move from one skill to another in a large-scale online curriculum.

A process model of simulated data showing how students use an LMS over time.

A Markov transition matrix that shows how students transition between problems.

Clusters of Curriculum Pacing patterns. Each cluster is a group of students who exhibited similar patterns of going through a math curriculum over time.

A Bayesian network that captures the some aspects of the math learning process.

Call for Short Papers

We are inviting short papers (up to 6 pages + references) that are related to the topics outlined below or on a topic that is aligned with the overall objective of the workshop.

Discovery and Analysis Methods

  • What methods are available to mine different representations of educational processes from sequential/temporal data?

  • How can we compare process representations of different individuals?

  • In what contexts some process representations work better than others?

  • How can we analyze process data with a high variance?

  • How can we visualize the learning and teaching processes?

Insights and Applications

  • How can we get insights about learning and teaching in the real world from process data?

  • How can we use process insights from data to inform real-world educational interventions?

  • How can we apply process models in learning engineering contexts to build applications like recommender systems and/or various types of adaptive learning systems?

Theory and Experimentation

  • How can we validate the discovered process models using psychology and cognitive science theory?

  • What type of experimental designs and infrastructure can help us discover more effective learning and teaching processes?

Tools and Data Management

  • Have you developed a tool or a software package to analyze the process data? How does your tool make process analysis easier?

  • What types of data storage formats are used in large-scale process data collection systems? What are their specific advantages?

Submission Guidelines

The submission deadline is 2 May 2021 Midnight Pacific Time.


We are inviting original and unpublished short research papers (up to 6 pages + references) for the workshop. The papers will be peer-reviewed by the committee members (see the list below).

Please download the paper templates from the EDM website: https://educationaldatamining.org/edm2021/submission/

Please send your anonymized short papers directly to nirmal@playpowerlabs.com. We will send you a submission acknowledgment as soon as possible. The identity of the sender will not influence the peer review process.


Acceptance notifications will be sent by 28 May 2021. Accepted papers will be published on CEUR-WS.

Program Committee

  • Nirmal Patel (Chief Data Scientist, Playpower Labs, Gandhinagar) [linkedin, researchgate]

  • Dr. Derek Lomas (Assistant Professor of Positive AI, Delft University of Technology, Delft) [university, linkedin, researchgate]

  • Dr. Agathe Merceron (Professor of Computer Science, Beuth University of Applied Sciences, Berlin) [university, google scholar]

  • Dr. Fabian Zehner (Post Doctoral Researcher, DIPF | Leibniz Institute for Research and Information in Education, Frankfurt) [university, google scholar]

  • Dr. Collin Lynch (Assistant Professor of Computer Science, North Carolina State University, Raleigh) [university, google scholar]

Venue

The workshop will be held alongside the 2021 Educational Data Mining Conference https://educationaldatamining.org/edm2021/

Registration note for K-12 teachers

K-12 Teachers from India: Registration is free for K-12 classroom teachers from India. Please contact Tirth Shah (tirth.shah@playpowerlabs.com) to get a separate registration form.

K-12 Teachers outside India: Please send your details to Tirth Shah (tirth.shah@playpowerlabs.com). We will share them with the conference organizers and request them to give you a discounted entry.

Contact

For any questions or comments, please reach out to Nirmal Patel (nirmal@playpowerlabs.com)

This workshop is being organized by Nirmal Patel, Derek Lomas, and Tirth Shah.