Data Science Projects

Applied Random Forest and Logistic Regression to predict Polycystic Ovary Syndrome (PCOS) from clinical data, achieving high accuracy and identifying key predictive factors like Follicle Number and Cycle regularity. This project highlights AI's potential in enhancing early diagnosis for better patient outcomes.

Analyzed multi-source data on global food production, population, and emissions using time series techniques. Developed and compared forecasting models (ARIMA, VAR) to predict future trends.

Leveraging machine learning on clinical data to build a robust model for early Chronic Kidney Disease (CKD) detection, improving patient outcomes and potentially reducing healthcare costs.

Leveraging Python, RDFLib, and standard medical ontologies (SNOMED-CT, LOINC) to transform tabular Chronic Kidney Disease data into a structured Knowledge Graph, highlighting data integration and problem-solving.

Discover how a custom big data system helps SPAR analyze promotional effectiveness, basket size, and cannibalization effects using Apache Spark and advanced analytics.