Accuracy rate is a measure of our prediction results, and it is a measure of how many of the predicted positive samples are actually positive. The recall rate is for our original sample, and it shows how many positive examples in the sample were correctly predicted. Accuracy = Predicted correctly/All If we want to be able to retrieve as much content as possible, this is the pursuit of "recall ratio", namely A/(A+C), the bigger the better. If we want to retrieve documents that we really want, that is, relevant as much as possible and irrelevant as little as possible, This is the pursuit of "accuracy", i.e. A/(A+B), the bigger the better.