educational-data-mining

Graph-based Knowledge Tracing: Modeling Student Proficiency Using Graph Neural Network

A library of algorithms for reproducing knowledge tracing, cognitive diagnosis, exercise recommendation and learning path recommendation models.

"Enhancing Knowledge Tracing via Adversarial Training", ACM MM 2021 (Oral).

Unsupervised Deep Autoencoders for Feature Extraction with Educational Data

Source code and data for the EDM 2022 paper

Agent-based simulation environment for Open & Distance Learning (ODL) research.

Python Data Mining Tool that analyzes grading patterns.

Dashboard for Corpus Procession and Exploration (currently German-only). Streamlit based for teaching and research: txt/xml corpora to csv-based datasets,  nlp (frequencies, tf-idf, cosinus), collocations, kwic, frequency curves, pos, semantic tagging, ner, pca, mds, t-sne, umap, dendrograms, topic-modeling, word2vec, document-term-topic-index

Regression Machine Learning Pipelines for predicting student performance

Python Data Mining Tool that analyzes Grading Patterns.

An end-to-end Data Science pipeline for analyzing mathematical misconceptions, featuring automated ETL, MySQL integration, and feature engineering for LLM training. Dockerized & CI/CD enabled.

PyTorch implementation of Attentive Knowledge Tracing for student performance prediction with AKT-PID and AKT-CID model variants.

PyTorch and TensorFlow implementations of LSTM, Transformers, and autoencoders for modeling sequential student behavior.

Python package for sequential pattern analysis using the GSP algorithm, supporting course sequencing and other applications.

A decision-support app that estimates whether a university student will drop out, stay enrolled, or graduate, and shows how early each prediction can be made.

SHAP-guided iterative feature engineering for educational data mining

Data scraping tool of Russian govt HE data

Applied ML research on student correctness prediction, adaptive learning, and educational data mining using EdNet.

A Genetic Algorithm to optimize textual features by their relevance, highlighting them in HTML.

AI-in-Education research prototype for detecting possible cognitive offloading through transparent behavioral analytics, verification, revision, and sensitivity analysis.