Sut Zaw Aung

Aspiring data analyst with strong foundation in Python, SQL, R, and Power BI. Proven experience in fraud detection, churn analysis, marketing response, and educational mobility projects. Eager to contribute as an intern or junior analyst; recognised for clear communication, collaboration, and translating data into actionable insights.

• Python, SQL, Jupyter Notebook, Logistic Regression, Chi‑Square Test, Proportion Z‑Test, Confidence Intervals, Cramér’s V.
• Quantitative analysis on survey data from 82 Myanmar students abroad; constructed Brain Drain Risk Index (78% high risk).
• Applied inferential statistics to test return intention proportions; association analysis revealed political/economic drivers.
• Visualised trends using Jupyter; produced policy‑focused insights on educational mobility behaviour.

• Python, Flask, HTML, CSS, PDF export, user authentication.
• Developed a Flask application to upload datasets, perform analysis, and generate downloadable PDF reports.
• Implemented user authentication, file management, and interactive prediction features.
• Built frontend interfaces and integrated backend processing for data workflows.

• R, SQL, feature engineering, stratified sampling, ROC curve, confusion matrix.
• Merged account, transaction, and campaign response data; built classification models to identify at‑risk customers.
• Evaluated model performance with ROC/AUC and confusion matrix metrics.

• R, Logistic Regression, XGBoost, SHAP, ROC/AUC.
• Used credit score, loan history, and fraud alert data for classification; visualised feature importance using SHAP.
• Generated ROC curves and feature importance plots for interpretability.

• R, Logistic Regression, Naive Bayes, XGBoost, funnel plots, heatmaps.
• Integrated customer, campaign, and response datasets for classification modeling.
• Visualised conversion and channel performance using funnel plots and heatmaps.

• Python, R, Isolation Forest, Random Forest, time‑series visualisation.
• Built supervised and unsupervised anomaly detection models using transaction and fraud alert data.
• Created dashboards with heatmaps and time‑series visualisations; applied OOP for modular analysis.