DBIS

In the ever-evolving landscape of education, the need for personalized and effective learning experiences is paramount. This thesis aims to address this challenge by developing a social bot tailored for self-regulated learning environments. The central focus lies in harnessing the power of xAPI statements to comprehensively track and interpret the learning progress of individuals engaging in self-regulated learning activities.

The study will build upon existing research in the fields of self-regulated learning, educational technology, and natural language processing. Relevant work includes the application of xAPI in e-learning environments, models for visualizing learning progress, and recommendation systems in educational contexts.

The first objective of the thesis is to explore how xAPI statements can be effectively employed to capture and analyze diverse learning activities. By leveraging xAPI, the social bot aims to provide a understanding of the learner’s journey, offering insights into strengths, weaknesses, and learning preferences.

Building on this foundation, the thesis delves into the implementation of the “GRETA – Kompetenzbilanz” model, a competency balance model designed to visualize learning progress. This model provides a holistic view of the learner’s competencies, allowing for a more nuanced and multifaceted representation of achievements. The challenge is to integrate this model seamlessly into the social bot, ensuring that learners can easily comprehend and engage with their progress in a meaningful way.

A crucial aspect of the social bot’s functionality is its ability to recommend learning materials based on the learner’s progress and preferences. This involves the exploration of recommendation algorithms, considering factors such as content relevance, difficulty level, and learning style. The goal is to enhance the learner’s experience by providing personalized content suggestions, fostering a more effective and engaging self-regulated learning journey.

Beyond tracking progress and recommending materials, the social bot aims to facilitate reflective practices. By incorporating features that encourage learners to reflect on their achievements, challenges, and strategies, the system aims to enhance metacognitive skills. This reflective component adds a layer of self-awareness to the learning process, empowering learners to adapt and optimize their strategies for continuous improvement.

Additionally, the thesis considers the incorporation of a frequently asked questions (FAQs) handling mechanism within the social bot. This feature aims to provide learners with quick and accurate information, addressing common queries and concerns in real-time. By doing so, the social bot becomes a comprehensive support system, not only for learning but also for clarifying doubts and fostering a positive learning environment.

Looking towards the future, the thesis explores the potential integration of MLOps (Machine Learning Operations) for the continuous improvement and retraining of recommendation models. This involves establishing a robust pipeline for model deployment, monitoring, and iterative refinement based on evolving learner feedback and educational content landscapes.

Prior work

• Affective Learning Recommendation Bot
• Utilize Mentoring Bots in Learning Management Systems
• Integration of a Moodle Chatbot into Online Courses Utilizing State-of-the-Art NLP Techniques

If you are interested in this thesis, a related topic or have additional questions, please do not hesitate to send a message to slupczynski@dbis.rwth-aachen.de and kissgen@dbis.rwth-aachen.de
Please apply with a meaningful CV and a recent transcript of your academic performance.