rpruthvikashyap / taxi_fair Goto Github PK

Ethical Considerations Request Review: Evaluate the ethical implications of alerting taxi drivers to customers who are unlikely to tip. Potential Bias: Consider the risk of biases in the data and its impact on certain customer groups. Objective Adjustment: Decide whether the model's objective should be adjusted for fairness. Driver Safety: Examine how the model's predictions might affect driver-customer interactions and safety. Feature Engineering Feature Selection: Choose relevant features such as ride distance, time of day, and payment method. Feature Extraction: Calculate new features like tip percentage based on provided data. Data Cleaning: Handle missing values and remove irrelevant or redundant features. Data Transformation: Convert categorical features into numerical format and standardize data. Modeling Random Forest: Model: Use GridSearchCV to tune a Random Forest classifier (rf) with hyperparameters such as max_depth, n_estimators, and min_samples_leaf. Scoring: Evaluate model performance using precision, recall, F1 score, and accuracy. Cross-validation: Perform cross-validation to validate model performance. Model Selection: Choose the best model based on the F1 score. XGBoost: Model: Use GridSearchCV to tune an XGBoost classifier (xgb) with parameters like max_depth, learning_rate, and n_estimators. Scoring: Evaluate model performance using the same metrics as above. Cross-validation: Perform cross-validation to validate model performance. Model Selection: Choose the best model based on the F1 score.