This repository stores mini data projects I have worked on.
Visualize spatial and temporal trend of business establishments in the heart of Melbourne.
Package used: pandas, plotly
Visualize closeness of different coffee blends using T-SNE, with data scraped from a local roaster's website.
Package used: pandas, numpy, plotly, sklearn, beautiful soup
Build and refine a default prediction model in an attempt to predict loan defaults. Also produced a gains table to evaluate the effectiveness of the model built.
Package used: pandas, numpy, sklearn, plotly, seaborn, optuna
Starbucks Optimization Strategy
Identify target customers to send promotions in a marketing campaign and evaluate effectiveness using custom objective functions.
Package used: pandas, numpy, lightgbm, plotly
An attempt to answer how fatalities on Victorian roads can be reduced. Dataset can be found here.
Tool used: Tableau
Visualize and analyze fictitious datasets (with further modifications) from DataCamp course Financial Analysis in Power BI.
PDF can be found here.
Tool used: Power BI Desktop
Visualize and analyze fictitious public datasets thelook_ecommerce hosted on GCP.
Tool used: BigQuery, Looker Studio