About the Workshop

Computing is an increasingly fundamental skill for students across disciplines. It enables them to solve complex, real and challenging problems and make a positive impact in the world. Yet, the field of computing education is still facing a range of problems from high failure and attrition rates, to challenges training and recruiting teachers, to the under-representation of women and students of color.

Advanced learning technologies, which use data and AI to improve student learning outcomes, have the potential to address these problems. However, the domain of CS education presents novel challenges for applying these techniques. CS presents domain-specific challenges, such as helping students effectively use tools like compilers and debuggers, and supporting complex, open-ended problems with many possible solutions. CS also presents unique opportunities for developing learning technologies, such as abundant and rich log data, including code traces that capture each detail of how students' solutions evolved over time. These domain-specific challenge and opportunities suggest the need for a specialized community of researchers, working at the intersection of AI, data-mining and computing education research.

The goal of the Educational Data Mining for Computer Science Education (CSEDM) workshop is to bring this community together to share insights for how to support and understand learning in the domain of CS using data. This field is nascent but growing, with research in computing education increasingly using data analysis approaches, and researchers in the EDM community increasing studying CS datasets. This workshop will help these researchers learn from each other, and develop the growing sub-field of CSEDM.

The theme of this workshop is the vision and future directions for the CSEDM community. In addition to submissions that share data mining approaches, methodologies, and experiences related to CSEDM, we encourage papers that critically review the current landscape of CSEDM research and practice and suggest fruitful future goals and visions. In this regard, we introduce a new track to this year's call, position track, accepting submissions that present a coherent discussion related to computer science educational data mining, including but not limited to diversity and equity, future research and practice directions, and impacts on CS education.

Topics of interest

• • • Deep learning approaches for analyzing, assessing, and scaffolding programming challenges

    • The role of big data in improving the analysis of educational programming data

    • Addressing special challenges of educational programming data with advanced machine learning approaches

    • Online learning environments for CS: implementation, design and best practices

    • Lessons learned for instructional design and support for learning

    • Evidence-based evaluation of interventions impacts on students' learning, motivation, and retention.

    • Discoveries in learning sciences based on learning data

• Predictive and descriptive modeling for CS courses

• Adaptation and personalization within CS learning environments

• Intelligent support for collaborative CS problem solving

• Machine learning approaches to analyze massive CS datasets and courses

• Affective, self-regulation, and motivational modeling of students as related to CS learning

• Adaptive feedback and adaptive assessment for CS learning

• Teaching approaches using AI tools

• Visual Learning Analytics and Dashboards for CS

• Recommender systems and in-course recommendations for CS learning

• Tool integration in learning management systems and educational organizations

• Discourse and dialogue research related to classroom, online, collaborative, or one-on-one learning of CS

• Social Network Analysis for CS learning environments

• Natural Language Processing for CS forums and discussions

• Specific studies of COVID impact on learning based on data analysis.

• CSEDM for inclusion, retention, and diversity

• Ethical considerations in CSEDM (e.g., privacy and ownership, Algorithmic and data bias)

• Classification of student program code

• Analysis of programming design and trajectory paths

• Feature extraction: identifying informative features from data for CSEDM

• Critical reviews of various aspects of CSEDM research and practice

• Vision and future research and practice directions for the CSEDM community

• Transparency, explainability, and trust in CSEDM research and practice

Submission Guidelines

We invite you to submit your original work for presentation and discussion. There will be two types of submissions:

• • Research Papers (4-8 pages) on AI and data mining being applied to computing education courses and data.

  • Position Papers or Work-in-progress Papers (4-6 pages) on:

    • Critical meta-reviews of CSEDM research and practice putting forward discussions of the vision and future research and practice directions for the CSEDM community.

    • Original, unpublished work-in-progress papers (incomplete or ongoing work, ready for feedback, but not yet fully developed).

  • Descriptions of CS Tools/Datasets/Infrastructure (2 pages), such as:

    • Descriptions of shareable Computer Science (CS) datasets

    • Descriptions of data mining / analytics approaches applied to specifically Computer Science datasets

    • Case studies of collaboration where reproducible practices were used to integrate or compose two or more data analysis tools from different teams

    • Descriptions of infrastructures that could collect and integrate data from multiple learning tools (e.g. forum posts, LMS activity and programming data)

Note: You will select your submission type on easychair.

Blinding: All submitted papers should be carefully blinded for review. Take care to remove all authors' names and identifying information (e.g. grant numbers), and refer to any of your prior work in the third person (e.g. "Previously, Smith et al. did ... [1]" rather than "In our prior work [1]").

*Note: references do not count towards the page limit. Authors can also include appendices to more clearly describe datasets and tools if necessary, and these do not count toward the page limit.

All submissions must be formatted using the EDM proceedings format (instead of LAK Companion format).

Submissions are handled via EasyChair: https://easychair.org/my/conference?conf=csedm23

Important Dates

• November, 2022: Open Call for Submissions

• December 16, 2022: Abstract Deadline for Papers from all tracks

• December 23, 2022: Paper Deadline for Papers from all tracks

• January 13, 2023: Notification of acceptance for Papers from all tracks

• February 3, 2023: Camera-Ready Version Deadline for Papers from all tracks

• March, 2022 7th CSEDM Workshop at LAK23

Publication

CSEDM proceedings will be published online via Zenodo. CSEDM proceedings should be considered semi-archival. This means that the papers are peer-reviewed, published and citable, but are still appropriate to extend for submission elsewhere. This is similar to the status of CHI extended abstracts. Some venues may expect that work published in a workshop should be revised before republication in a conference or journal, and we expect this should be straightforward to do with feedback you get by presenting at CSEDM. For example, ACM conferences expect an extension of 25%. Authors retain the copyright to their work when published through Zenodo at CSEDM.