Hong Huang

With the rapid advancement of the civil aviation sector and the concurrent expansion of pilot training programs, a pressing need arises for more efficient assessment methodologies during the pilot training process. Traditional written evaluations conducted by flight instructors are often marred by subjectivity and inefficiency, rendering them inadequate to satisfy the stringent demands of Competency-Based Training and Assessment (CBTA) frameworks. To address this challenge, this study presents a novel multi-label classification model that seamlessly integrates RoBERTa, a robust language model, with Graph Convolutional Networks (GCNs). By simultaneously modeling text features and label interdependencies, this model enables the automated, multi-dimensional classification of instructor evaluations. It incorporates a dynamic weight fusion strategy, which intelligently adjusts the output weights of RoBERTa and GCNs based on label correlations. Additionally, it introduces a label co-occurrence graph convolution layer, designed to capture intricate higher-order dependencies among labels. This study is based on a real-world dataset comprising 1078 evaluations and 158 labels, covering six major dimensions, including operational capabilities and communication skills. To provide context for the improvement, the proposed RoBERTa + GCN model is compared with key baseline models, such as BERT and LSTM. The results show that the RoBERTa + GCN model achieves an F1 score of 0.9737, representing an average improvement of 4.73% over these traditional methods. This approach enhances the consistency and efficiency of flight training assessments and provides new insights into integrating natural language processing and graph neural networks, demonstrating broad application prospects.

This paper proposes a new model of collaborative filtering that introduces a user difference factor to address the issue of less pronounced similarity performance in traditional algorithms. The algorithm is applied to the recommendation of pilots’ core competency behavior indicators to support pilots’ daily training arrangements, improve their core competencies, and ensure civil aviation flight safety. Firstly, based on traditional collaborative filtering methods, a user difference factor is introduced to improve the Pearson similarity calculation model. Secondly, the advantages and disadvantages of collaborative filtering recommendation models were evaluated using various methods such as average absolute error, accuracy, recall, and diversity. Finally, the new model is applied to the recommendation of PLM core competency behavior indicators, providing a recommendation list of different pilot behavior indicators to support their rehabilitation or enhanced training plans and arrangements. The calculation results show that the new model of collaborative filtering demonstrates better advantages, not only reducing the MAE value, but also improving the accuracy, recall, and diversity of the calculation results, providing effective guidance and theoretical support for pilots’ flight training and safe flight.