Clustering to Detect Fabricated Survey Answers
Research project implementing GMM, One-class SVM and IForest to detect falsified answers in a national Dominican Republic household survey (ENGIH 2018).
Projects
The Happiness-Suicide Connection
Set of Tableau dashboards exploring the happiness index and suicide rates globally.
Logistic Regression to predict Diabetes
Flask deployment of logistic regression classifier to predict whether a person may have diabetes or not using clinical measurements.
Stable Matching in the Housing Market
Gale-Shapley Algorithm to match tenants and landlords according to their preferences in Python.
Neural Network and Logistic Regression from scratch
Logistic Regression and Neural Network models built from scratch to compare performance on classifying diabetes outcomes
Currently working on...
- Flask deployment of patient-level platform to predict heart diseases using a machine learning approach in Python
- Machine learning classifier for success of bank marketing campaign in Python
- Calculating expected returns of stock portfolio using the Capital Asset Pricing Model (CAPM)