Arm Kantawee

Churn Modeling in Telco Business

Developed a predictive model using XGBoost to identify at-risk customers in a telecommunications dataset, achieving an AUC-ROC of 0.88.

View Code

Stock Price Prediction via Regression

Applied multiple linear regression and LSTM neural networks to forecast short-term stock price movements using historical data and technical indicators.

View Code

Customer Segmentation Analysis

Performed K-Means and Hierarchical clustering on retail transaction data to identify distinct customer personas and optimize marketing strategies.

View Code