Open-Source Adaptive Learning Framework (ALF) for STEM Education
The author has built an open-source, modular, and bilingual Adaptive Learning Framework (ALF) for STEM education, with a focus on transparency and extensibility.
Why it matters
This open-source framework can empower educators and researchers to build and experiment with adaptive learning systems in a transparent and collaborative manner.
Key Points
- 1ALF models a simple but powerful adaptive learning loop: Diagnosis → Drill → Integration
- 2It uses clean JSON modules that anyone can write, with no black boxes or hidden heuristics
- 3The framework includes a JSON Problem Bank, an Adaptive Learner state machine, and a thin Engine Layer
- 4The Streamlit-based UI supports English and Dutch, with the logic residing in the engine
Details
ALF is a lightweight, open-source framework that aims to make adaptive learning accessible and transparent. It is built with a modular design, where each topic is defined in a standalone JSON file containing the question, correct answer, common error patterns, drill prompts, and integration test. This makes the framework easy to extend, as educators can add new topics without modifying the core engine. The Adaptive Learner is a simple, readable Python class that moves through the diagnosis, drill, and integration phases, storing the learner's history and current state. The thin Engine Layer orchestrates the learner initialization, answer routing, and structured result reporting to the UI. The Streamlit-based UI supports both English and Dutch, with the focus on keeping the interface minimal and the logic in the engine. The author's goal is to make adaptive learning open, transparent, and accessible to educators, developers, and researchers, rather than locked behind corporate walls.
No comments yet
Be the first to comment