AIStudyBuddy
AI-based Support for Study Planning
The joint application "AIStudyBuddy – AI-based support for study planning" was submitted with RWTH as the applicant university within the framework of the Federal-Länder Initiative for the Promotion of Artificial Intelligence in Higher Education. Together with RWTH, Ruhr University Bochum (RUB) and the University of Wuppertal (BUW) are working on using modern AI technologies to support the planning and reflection of individual study paths.
StudyBuddy for Students
Students are provided with StudyBuddy, a tool for informed and evidence-based study planning over multiple semesters into the future. Study Buddy provides a graphical representation of study progress and provides actionable feedback.
This is based on rule-based study progression plans as well as progression profiles determined by AI technology that lead to successful degree completions. Generic study plans are thus supplemented by a tool for individual study planning, which is continuously adapted, justified, and reflected upon.
BuddyAnalytics for Degree Program Designers
BuddyAnalytics provides program designers with an interactive tool that supports planning decisions such as competency-based curriculum development and student advising.
By analyzing and visualizing study program data from different higher education systems, adjustments and improvements to study programs can be developed in an evidence-based manner. In doing so, problems in the design of study programs and study behavior deviating from the course of study planning can be identified.
The AIStudyBuddy Project Uses Modern AI Technologies
The project combines two AI paradigms for this purpose: data-driven through process mining and rule-based through Answer Set Programming (ASP). Both components are part of a reference architecture that follows principles such as ethics by design and privacy preservation.
Process Mining / Machine Learning
Process mining is used to analyze study behavior based on data from campus, learning management, and examination systems. It contrasts real study trajectories with intended ones.
Answer Set Programming – ASP
Answer Set Programming is used to transform examination regulations and other rules into a model of rules and constraints in order to generate transparent rationales for feedback in study planning that is comprehensible to non-domain experts.
Goals of the Project
The strategic goal is the standardized exchange of student data between universities as a first step toward cross-university student monitoring. The project pays particular attention to the determinants of acceptance in the introduction of AI-based assistive technologies.
As a result, the project network, supported by Prof. Dr. Ulrik Schroeder, PD Dr. Malte Persike, Prof. Dr. Gerhard Lakemeyer from RWTH, Prof. Dr. Wil van der Aalst from RWTH, Prof. Dr. Maren Scheffel from RUB, Prof. Dr. Sebastian Weydner-Volkmann from RUB, Dr. Peter Salden from RUB, Prof. Dr. Kerstin Schneider from BUW, and Dr. Simon Görtz from BUW, use evidence-based study monitoring, interactive tools for course planning, and data-driven curriculum design to enable even more successful courses of study and graduates.
Goals
- Learning – Through automatic reviews and individual recommendations, study plans can be designed and optimized.
- Teaching – Study plans, examination regulations, and the organization of studies can be adapted and improved by identifying problems.
- Application – Support of individual planning and reflection of study progressions by combining machine learning and answer set programming.
- Network – Sharing approaches and experiences. Evidence-based study monitoring, interactive tools for course planning, and data-driven curriculum design will be used to enable even more successful study courses and graduates.