Sufeeya Parween

Developed a machine learning model with over 85% accuracy using 5,000+ soil and climate records to recommend optimal crops. Utilized Python libraries for data preprocessing, analysis, and visualization, building a real-time, location-based interface to improve crop yield predictions by up to 20%.

Developed an end-to-end machine learning pipeline to predict diabetes, achieving 81% test accuracy and reducing model training time by 30% through feature optimization. Designed and deployed a Flask-based web application with a responsive UI, enabling real-time single-patient predictions and integrating the trained model using Pickle for instant RESTful API responses.

Developed a real-time ML pipeline to detect faulty semiconductor wafers using sensor data, achieving approximately 98% accuracy with XGBoost and handling class imbalance. Deployed a prediction API using Flask and Uvicorn with MongoDB for scalable data handling, significantly reducing manual inspection costs and potential downtime.

Conducted comprehensive statistical analysis on Algerian forest fire data to identify environmental patterns and correlations with fire events. Improved feature relevance by 50% through statistical analysis and delivered clear visual insights through customized charts and heatmaps, enhancing interpretability for stakeholders.

Studied core concepts of computer applications, including programming languages, software development, and database management. Gained practical skills in various computing domains relevant to modern IT environments.