Transforming Dining-Centric Credit Cards: Seizing the Shopping Revolution

This project aimed to understand customer behavior by thoroughly analyzing credit card activities and customer transactions. Using RFM segmentation and K-Means Clustering, various types of customers were uncovered, and recommendations were given to boost customer satisfaction and business growth.

Uncovering the Silent Epidemic: Insights into Stroke Risk Factors

Our group developed a model predicting the risk of stroke in patients. We employed imbalanced data techniques like SMOTE and ADASYN, and classification models like logistic regression and tree-based models to achieve our objective. The key features in the prediction were also identified using SHAP.

SB19: Bringing Pinoy Pop to the World

This project aimed to promote SB19's music by helping more people discover their sound. After uncovering patterns in their music performance through time-series analysis, we developed a Spotify recommender engine, which allows users to input a top Philippine hit and recommends similar SB19 songs.

Language of Engineering Careers: U.S. LinkedIn Job Description Analysis

This project leveraged NLP techniques and GenAI to analyze a LinkedIn dataset containing information on job posts. By employing topic modeling and retrieval-augmented generation, a chatbot was created that provides relevant job posts and descriptions according to the skills entered by the user.

Watts Up, 'Pinas? Illuminating Insights in the Electricity Spot Market with K-Means Clustering and Time Series Analysis

This project aims to aid the DOE in its overall goal of achieving net zero and reducing electricity costs. K-Means Clustering revealed diverse bidding patterns of power plants that affect electricity pricing. The availability of renewable energy was also assessed with time-series forecasting.