Internet traffic monitoring and traffic characterization are essential for managing and optimizing network infrastructures. In general, the traffic classifier based on Machine Learning (ML) might not perform well for some cases such as imbalanced data sets. To address this issue, we propose an evaluation measure aiming for the case of imbalance multi-class network traffic classification based on the multi-objective metric, the area under the ROC (Receiver Operating Characteristic) curve, or simply AUC. The experiments show that our method is more sensitive in the case that the Minority classes being misclassified into the Majority classes than the otherwise. Hence, we could use this measure to evaluate the classifier performance in the distinguishing of misclassifying the Minority applications into the Majority applications.