Data Science Fellow
Current➤Project: “Develop a multi-class classification model to identify pancreatic cancer”● Created and optimized 6 different supervised algorithms in a multi-class classification campaign, including Logistic Regression, KNN, Decision Tree, Random Forest, Gradient Boosting, and Support Vector Machine Classifiers, achieving high performance metrics (accuracy: 78.8%, f1-score: 78.7%, log-loss: 58.2%).● Innovated and implemented an effective Support Vector Machine Classifier for early non-invasive cancer detection, rigorously evaluating model accuracy using a confusion matrix and prioritizing enhanced ROI for healthcare providers.➤Project: “Time series forecasting using ARIMA and PyCaret”● Performed thorough time series analysis and developed two predictive models (Seasonal ARIMA, Exponential Smoothing) using PyCaret that effectively leveraged historical data to make realistic forecasting.● Empowered strategic advertising decisions and ensured sustainable tourism development for the Tourism Bureau by creating a robust tool to understand past trends, predict future scenarios, and facilitate data-driven decision-making.