- π Final-year Information Technology (Data Science) student at Ho Chi Minh City University of Industry and Trade (HUIT)
- π Focused on the end-to-end analytics workflow: EDA β feature engineering β machine learning β Power BI dashboards
- π§ Comfortable across the stack: Python, SQL, Power BI, PyTorch, Scikit-learn
- π± Currently exploring deep learning applications and deployment
- π« Reach me at datnguyen23062005@gmail.com
End-to-end deepfake video detection system combining EfficientNet-B4 with a Transformer temporal head.
- Unified corpus of 13,692 processed face images from FaceForensics++, DFDC, and Celeb-DF
- MediaPipe BlazeFace face-cropping, multi-frame classification, focal loss training
- 0.977 validation ROC-AUC, deployed as an interactive Flask web demo
PythonPyTorchEfficientNet-B4Transformer
Predictive analytics pipeline on 3 million U.S. domestic flight records (2019β2023).
- Feature engineering, EDA, and clustering to uncover delay patterns across airlines, airports, and seasons
- Random Forest, XGBoost, and LightGBM models with SHAP-based interpretability
- Interactive Power BI dashboard for delay trends and geographic visualization
PythonPandasScikit-learnXGBoostLightGBMSHAPPower BI
Ho Chi Minh City University of Industry and Trade (HUIT) Bachelor of Information Technology β Data Science | 2023 β 2027 |

