Lily I want to understand what data imbalance means in machine learning. How does an uneven distribution of classes affect model performance and predictions? Can someone also explain common techniques used to handle imbalanced datasets?