epfl-ml4ed/ripple
Interpretability on raw time series with graph neural nets and concept activation vectors. Featured at AAAI 2023.
This project helps educators predict student performance early in online courses without needing to manually create complex features from student interaction data. It takes raw student clickstream data from MOOCs and outputs predictions of student pass/fail status, along with interpretable insights into why a student might succeed or struggle. Educators, instructional designers, and educational researchers would use this.
No commits in the last 6 months.
Use this if you need to predict student outcomes in online courses based on raw interaction data and want to understand the underlying reasons for their performance to design targeted interventions.
Not ideal if your data is not time-series based or if you already have highly refined, expert-designed features that perform well.
Stars
8
Forks
2
Language
Jupyter Notebook
License
MIT
Last pushed
Jan 25, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/epfl-ml4ed/ripple"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
obss/sahi
Framework agnostic sliced/tiled inference + interactive ui + error analysis plots
tensorflow/tcav
Code for the TCAV ML interpretability project
MAIF/shapash
🔅 Shapash: User-friendly Explainability and Interpretability to Develop Reliable and Transparent...
TeamHG-Memex/eli5
A library for debugging/inspecting machine learning classifiers and explaining their predictions
csinva/imodels
Interpretable ML package 🔍 for concise, transparent, and accurate predictive modeling...