飞行安全是飞行运输中的一个重要课题,对飞行纪录数据进行分析是判明飞行事故原因的重要依据,所以对飞行记录数据进行数据挖掘则成为一种有效的飞行监控技术。该文针对具有高维特征和大样本数据集的飞行纪录的飞行纪录数据学习问题,提出了一种新型的判别异常飞行事件的特征提取方法,并通过实验表明该方法对某一机型的飞行纪录数据取得了良好的实验结果。 Flight safety is an important issue in flight transportation. Analysis of flight record data is an important basis for determining the cause of flight accidents. Therefore, data mining of flight record data becomes an effective flight monitoring technology. Aiming at the problem of flight record data learning of flight records with high-dimensional features and large sample datasets, this paper proposes a novel feature extraction method for discriminating anomalous flight events. Experiments show that this method is effective for a certain type of flight Record data achieved good experimental results.