Mehzabeen Gheesah

Dynamic data scientist with a 16-year background in education, possessing a unique combination of data analytics skills and an innate ability to communicate and educate. Adept at using machine learning and data visualization techniques to drive insights, combined with strong interpersonal skills, ensuring data-driven solutions are accessible and actionable for all stakeholders.

Final Project-Disease Prediction

• Tools - Python, Pandas, Numpy, Matplotlib, Seaborn, Natural Language Processing Toolkit, SciKit-Learn
• Algorithms - Linear Support Vector Classifier, Random Forest Classifier,
• Objective - Use text data to predict a disease,
• Result - The RandomForest Classifier demonstrated superior performance with higher accuracy and fewer misclassification. This model was then locally deployed, leading to the development of a user-friendly app to diagnose diseases based on symptoms

Machine Learning - Unsupervised,

• Tools - Python, Scikit-Learn, Pandas, Numpy, Matplotlib, Seaborn, Jupyter Notebook, Git, GitHub
• Algorithms - Principal Component Analysis (PCA), KMeans clustering, Hierarchical Clustering,
• Objective - The main objective is to segment the customers in such a way that the distributor can devise targeted marketing and service strategies for each segment, thus improving business performance and customer satisfaction.
• Result - I discerned three distinct customer segments based on their purchasing behaviors, each correlating to different types of establishments and regions, thereby enabling the potential for tailored marketing strategies

Machine Learning - Supervised

• Tools - Python, Scikit-Learn, Pandas, Numpy, Matplotlib, Seaborn
• Algorithms - Logistic Regression, Random Forest Classifier, XGBoost
• Objective - The goal of this project is to use supervised learning techniques to build a machine-learning model that can predict whether a patient has diabetes or not, based on certain diagnostic measurements
• Result - I was able to accurately predict the onset of diabetes with the highest accuracy and precision using the XGBoost model, demonstrating its effectiveness in this health-focused machine learning application