siam29/Hybrid-Feature-Engineering-and-Ensemble-Learning siam29/Hybrid-Feature-Engineering-and-Ensemble-Learning

Twitter logo Share LinkedIn logo Share Facebook logo Share
siam29/Hybrid-Feature-Engineering-and-Ensemble-Learning
github

siam29

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

In this ML project, I proposed a methodology that provided an outperformed performance compared to another existing paper. For the comparison here focused mainly on F1, accuracy, AUC, and ROC score. This methodology provides a 99.96% accuracy score and 90.05% F1 score.Β 

More Repositories

1

The-Third-Eye-Smart-Glasses-for-the-Blind-without-Arduino

Implemented a smart glasses project for the visually impaired, using advanced IoT technologies without Arduino. The system offers real-time feedback and navigation assistance, with comprehensive documentation and code available. Aimed at enhancing accessibility and independence.
2
star
2

Exploring-Explainable-AI-Demystifying-DT-RF-KNN-XGBC

Implemented XAI techniques to enhance transparency in fraud detection models. I employed techniques such as SHAP, LIME on DT, RF, XGBC, and KNN to offer lucid explanations for transactions that were flagged.
Jupyter Notebook
1
star
3

Ensemble-majority-voting-Hard

In this project, we implemented an ensemble learning approach using majority voting (hard voting) with five machine learning classifiers: DT, RF, XGBC, ANN, and KNN. The ensemble model achieved an impressive accuracy score of 99.95% and an F1 score of 85.51%.
Jupyter Notebook
1
star
4

Credit-Card-Fraud-Detection-in-Real-Time

This project delivers a fast and efficient fraud detection methodology, providing predictions in under a second, emphasizing the importance of both high performance and quick response times.
Jupyter Notebook
1
star