Status: In progress
A planning-stage research prototype exploring agentic AI, RAG, and knowledge graphs for personalized university and major recommendations in Cambodia.
Completed:
End-to-end customer segmentation using PCA and K-Means clustering to identify distinct customer segments and support data-driven marketing strategies.
Completed:
Statistical analysis of categorical datasets using χ²-test and Correspondence Analysis to uncover hidden relationships.
Completed:
Sequence-to-sequence LSTM model with attention mechanism to transfer Khmer text between linguistic registers (normal to royal) while preserving semantic content.
Completed:
Designed automated Python ETL pipelines, modeled PostgreSQL data warehouses, and developed interactive Power BI dashboards for economic data analysis.
Completed:
Machine learning classification model to detect fraudulent banking transactions using ensemble methods and SMOTE.
Completed:
NLP-powered spam email detection using machine learning with TF-IDF vectorization and six supervised learning classifiers.