Data plays a pivotal role in the aviation industry, driving advancements in safety, operational efficiency, and passenger experience. It is an important and interesting topic for the classification of imbalanced data. There exist many methods to classify the imbalanced data, but most of them only utilize the labeled patterns for classification tasks. When the labeled patterns in minority classes are extremely rare, it is quite difficult to achieve satisfactory classification results. To deal with this problem, we take full advantage of the information in whole data to perform classification tasks. Firstly, to enrich the distribution of the minority class, label propagation algorithm is used for assigning pseudo labels to the test set data. Second, the samples in the test set with pseudo labels as positive examples are combined with the training set samples to form a new training set, and SMOTE-ENN approach is employed for resampling. Finally, the classifier is trained using the resampled dataset. The experiment is conducted using multiple datasets from KEEL for validation, and the results show that our method outperforms other sampling methods in terms of AUC and G-mean metrics for classifying imbalanced data.