Portfolio
Exploratory risk analysis dashboard
I used Power BI to explore how national workplace risk statistics vary depending on inspection program and US region. I found that state programs and the US West associated with safer working environments in 2012.
Title: 2012 National Workplace risk data
Links:
description | GitHub
demo | R Shiny app
Topics:
Exploratory analysis | Risk analysis | Dashboards
Software:
Power BI | R | Shiny | Quarto
Supervised classification machine learning
I used Python to build supervised classification machine learning models with the bank marketing dataset to classify whether a client will subscribe a term account. I progressed three different model types through optimization techniques and deployed a best-performing model.
Title: Classifying term subscription with client banking data
Links:
description & notebook | GitHub
deployed model (demo) | R Shiny app
Topics:
Supervised classification | Preprocessing | Modeling
Software:
Python | R | Shiny | Quarto
Predicting ICU need during the COVID-19 pandemic
I used Python modules (sklearn, pandas, numpy) to conduct EDA of the COVID-19 ICU dataset and build predictive models to classify whether a person seeking treatment for COVID will require the ICU. I found that a random forest classifier had the best classification performance with an accuracy near 92%.
Links:
notebook | GitHub
Topics:
Supervised classification | Preprocessing | Modeling
Software:
Python | R | Quarto