I'm new to machine learning and struggle to interpret the results I get from different measures of performance. If for several prediction models I have e.g. accuracy, precision, recall, F1, FPR, and MCC and want to find out which model performs best, what do I look for? I would assume accuracy is most important?
Also, what influence does the error have on the interpretation? E.g. method 1 has an accuracy of 87 +- 5% and method 2 has 84 +- 5% I would assume method 1 to be better. Is that correct?