Projects
Welcome to my portfolio! Here, I showcase my best work to give you a glimpse of my skills and experience. Feel free to explore and get in touch if you have any questions or opportunities to collaborate.
Blinkit Data Analysis Dashboard
Designed and developed an interactive Power BI dashboard to analyze sales and performance insights using the Blinkit dataset..
-
Built data visualizations including bar charts, pie charts, and KPIs to highlight key business trends..
-
Applied DAX functions for calculations such as totals and percentages to enhance analytical depth
-
Streamlined insights delivery by connecting Excel as the primary data source, ensuring accessibility and scalability.
Credit EDA (Exploratory Data Analysis)
Led an Exploratory Data Analysis (EDA) on credit-related datasets, including cleaning and preprocessing steps to manage missing data, detect and treat outliers, and standardize variable formats (e.g., converting negative "DAYS" values to positive).
-
Conducted EDA on credit dataset with preprocessing, handling missing values, and outlier treatment.
-
Performed statistical analysis and visualizations (Pandas, Matplotlib, Seaborn) to identify feature correlations and distributions.
-
Derived key insights on borrower profiles, credit amounts, and risk factors, supporting better credit risk assessment.
Predicting Mortality of Heart Failure Patients
Conducted data preprocessing and EDA on heart failure patient records, addressing missing data, feature engineering, and class imbalance
-
Built and benchmarked machine-learning models—including Random Forest, XGBoost, LGBM, and K-Nearest Neighbors—to predict in-hospital mortality.
-
Utilized a stacking ensemble approach to combine model strengths, achieving top predictive performance.
-
Documented precise evaluation metrics (AUC, sensitivity, specificity) for training and testing phases to quantify model reliability and clinical relevance.
