LeelaParhyar/Typhoid-Analysis-in-Uganda-using-Regression-Techniques.

This project analyzes and predicts the number of Typhoid cases in the different districts of Uganda using the following, regression models, Decision Tree, Random Forest and Support Vector Regression. # Typhoid Analysis in Uganda using Regression Techniques. This project analyzes and predicts the patterns and occurrences of the number of Typhoid cases in the different districts of Uganda using the following, regression models (Decision Tree, Random Forest and Support Vector Regression). ## Importing Necessary Libraries. Import esential libraries for data analysis, data manipulation, visualisation and model training. ## Dataset: Loading a dataset that was collected by the Uganda National Meterrological Authority and the Ministry of Health, contains historical and enviromental data about Typhoid in differnt districts of Uganda. ## Preprocessing: preprocess and Clean the dataset that contains historical typhoid outbreaks, by handling missing values, droping and moving columns. ## Feature Selection. Split data into features (X) and target variable (y), Select some of the most important features to improve model activity. ## Feature Scaling. Feature scaling is done in order to normalized the data to improve model accuracy. ## Model Selection: In this process Regression techniques were selected due to continuous nature of data, and the models chosen include Decision Tree, Random Forest, and Support Vector Regression (SVR) ## Hyper Parameter Tuning. Hyperparameters are used for tuning in order to inprove the performance model. ## Model Testing. Outputs a mean Squared Error (MSE), Mean Absolute Error (MAE) and R-squared (R2) Score to evaluate performance.