Introduction: Developing a genetic risk score model to predict Type 2 Diabetes onset using demographic and medical data from the Pima Indian population. Aim to identify high-risk individuals for early intervention.
Dataset Explanation: Utilized the NIDDK dataset from the Diabetes Prevention Program, containing medical and lifestyle data of over 3,000 participants, focusing on features like pregnancies, glucose levels, BMI, and more.
Methodology & Implementation: Processed data with Pandas, imputed missing values using KNN, removed outliers, and visualized data relationships. Utilized Spark ML for model training with Decision Tree Classifier.
Metric Evaluation & Results: Achieved 73% accuracy with Decision Tree Classifier, demonstrating promising precision, recall, and F1 score. Confusion matrix and classification report validate model effectiveness.
Conclusion: In conclusion, this project analyzed and predicted diabetes outcomes using a cleaned dataset, leveraging KNN imputation and outlier removal. EDA provided insights, and a decision tree classifier trained with PySpark achieved 73% accuracy. These findings inform diabetes risk assessment, with potential for further enhancement through algorithm exploration and feature incorporation.