I am building a binary classification (positive/negative) model and use the typical threshold value of 0.5, meaning that if the output probability is above 0.5 the model predicts a positive, and if it below 0.5 the model predicts a negative. What would happen to the recall of my model if I reduced my threshold to 0.3? Recall would go down because I would classify less observations as positive and therefore would correctly identify less of the positive cases Recall would go up because I would classify more observations as positives and therefore would correctly identify more of the positive cases Recall would not change because it does not depend on the threshold value Recall would go up because more of the predicted positives were actually positives 1 point