Humanizing
Data
This page contains selected applied data science projects developed as part of my formal training through MIT’s Applied Data Science & AI program.
​
The work reflects my focus on model evaluation, data mining, feature engineering, and translating analytical findings into practical recommendations—particularly in organizational, workforce, and decision-making contexts.
​
While I do not position myself as a production data scientist, this experience informs my work in AI governance, evaluation, and responsible deployment by enabling informed dialogue with technical teams and critical assessment of model behavior, tradeoffs, and impact.
​
Selected projects below link to full Jupyter notebooks, including code, analysis, and conclusions.
Food Delivery Aggregator
Skills: Exploratory Data Analysis & Visualization, Python, Numpy, Pandas, Matplotlib, Seaborn, Summary & Recommendations
​
Predict Patients' Length-of-Stay upon Admission
Skills: EDA & Visualization, k-NN Imputation, Linear Regression Modeling, Statistical Analysis (Pearson Correlation Coefficient, RMSE, MAE, R2 & MAPE)
Determine Employee Risk of Attrition
Skills: EDA & Visualization, Logistic Regression Modeling, Decision Tree Classifier, Random Forest Classifier, K-NN, GridSearchCV Hyperparameter Tuning, model evaluations and comparison
Detect Patterns in Player Performances
Skills: EDA & Visualization, Cluster Analyses: PCA, K-Means, K-Mediods, Hierarchical, Gaussian Mixture Model, DBSCAN), Python, Numpy, Pandas, Matplotlib, Seaborn, Sklearn
Skills: Regression Models - Decision Tree Regressor, Bagging, Random Forest Regressor, AdaBoost, Gradient Boost, XGBoost, model evaluations and comparison
Identify Defaulter Profile & Credit Risk Features
Skills: EDA & Visualization, Feature Engineering, Logistic Regression Modeling, Decision Tree Classifier, Random Forest Classifier, Hyperparameter Tuning, model evaluations and comparison, recommendations for deployment
