Measure identification of classifier performance

This paper analyzes the current wide-used measure identification of classifier performance--accuracy and error rates. However, in unbalanced data set, semantic-related multi-class, different costs for different misclassification type and other classification problems, there are many defects when accuracy and error rates are used to measure the classifier performance. In order to solve the above problems, precision, recall, mistake, omitting F-measure ratio and classification cost matrix, loss function are integrated used to measure the performance of classifier.