Balancing Efficiency vs. Effectiveness and Providing Missing Label Robustness in Multi-Label Stream Classification
-
Updated
Feb 7, 2024 - Python
Balancing Efficiency vs. Effectiveness and Providing Missing Label Robustness in Multi-Label Stream Classification
Credit Card Fraud Detection
A short research paper that investigates cheap frame filtering techniques to predict model drift in neural networks
Detection and classification of anomalous events in oil extraction. Incremental learning methods applied to the Petrobras 3W dataset.
Enhancing electricity price forecasting accuracy using a hybrid model combining GRU and XGBoost with detection-informed retraining for concept drift.
EP2420 Course project. Part 1 is for warming up. Part 2 is about online learning.
"Past performance of machine learning model is no guarantee of future results." We call it "model drift" or "model decay". This repository will introduce various methods for detecting model drift.
Federated Learning on Multi-label Evolving Data Streams
Music album popularity prediction classic ML model showcasting MLOps, versioning, feature selection, cross valdiation and concept drift.
Algorithms proposed in the following master dissertation: OLIVEIRA, Gustavo Henrique Ferreira de Miranda. Previsão de séries temporais na presença de mudança de conceito: uma abordagem baseada em PSO. 2018. Dissertação de Mestrado. Universidade Federal de Pernambuco.
PhD Research of an approach to deal with concept drift in process mining using transition matrices
The implementation of Synthetic Minority Oversampling based on stream Clustering (SMOClust)
EDIST2: Error Distance Approach for Drift Detection and Monitoring
A classifier for heterogeneous concept drift inspired in the biologically memory model.
Master Thesis entitled "Lightweight Real-Time Feature Monitoring"
Coding tasks regarding different machine learning models, their use within a Flask-API and concept drift detection for the lecture Artificial Intelligence in Service Systems (AISS) at Karlsruhe Institute of Technology, winter term 2020/2021.
Learning High-Dimensional Evolving Data Streams With Limited Labels
Light weight hyperparameter tuning for streaming scenarios
Add a description, image, and links to the concept-drift topic page so that developers can more easily learn about it.
To associate your repository with the concept-drift topic, visit your repo's landing page and select "manage topics."