Development of a Sepsis Prediction Model Using Electronic Health Records
I developed a model to predict hospital-acquired sepsis in patients. I obtained the electronic health records from the MIMIC-III database and created new data frames using SQLite.
sqlite, python 3, pandas, descriptive statistics, matplotlib, seaborn, scikit-learn, random forest classifier, xgboost
Predicting Chronic Kidney Disease
In this study, I tested the efficacy of three classification models in predicting patients with chronic kidney disease. Data was obtained from the University of California, Irvine’s Machine Learning Data Repository.
python 3, pandas, descriptive statistics, seaborn, holoviews, hvplot, bokeh, scikit-learn, decision tree, random forest classifier, support vector machine
Changes in Inequality in Houston, TX
I compared the educational attainment level of people living below the poverty line and those receiving an income of $200,000 or more annually.
python 3, pandas, matplotlib, descriptive statistics
Efficacy of Gun Control Laws
I looked at the number of people affected in mass shootings and the number of these events between January 1, 2013 and August 13, 2019. I also proposed hypotheses, which could be tested if new gun control laws were enacted.
google sheets, hypothesis testing