Dual-Paradigm Binary Malware Detection using Batch Machine Learning and CapyMOA Streaming | SHAP | Active Learning | Concept Drift
-
Updated
Sep 1, 2026 - Jupyter Notebook
Dual-Paradigm Binary Malware Detection using Batch Machine Learning and CapyMOA Streaming | SHAP | Active Learning | Concept Drift
Hoeffding Adaptive Tree Regressor for CapyMOA with MOA implementation, wrapper, and streaming regression benchmarks.
Online forecasting of bus passenger demand using CapyMOA, with autoregressive models and a station recommendation prototype. Academic team research at the University of Waikato using public transport data from Salvador, Brazil.
Memory-bounded streaming ensemble (Online Bagging + ADWIN + Leveraged Bagging) with a Hoeffding-Tree champion, including a documented post-mortem on a MOA Java-heap memory-measurement bug and its fix.
Lab on Online Machine Learning for M2DS Ecole Polytechnique Data Stream Processing course.
Six-person COMPX523 group project: online peak-hour boundary detection and drift tests on three months of historical estimated boarding counts for 10 Salvador bus stops, replayed as a stream.
Online kNN for evolving data streams (COMPX523 coursework): three windowing strategies (global FIFO, per-class, age-weighted) on CapyMOA/MOA.
To associate your repository with the capymoa topic, visit your repo's landing page and select "manage topics."