Machine learning-based anomaly detection and prediction in commercial aircraft using autonomous surveillance data.

Regarding the transportation of people, commodities, and other items, aeroplanes are an essential need for society. Despite the generally low danger associated with various modes of transportation, some accidents may occur. The creation of a machine learning model employing data from autonomous-reliant surveillance transmissions is essential for the detection and prediction of commercial aircraft accidents. This research included the development of abnormal categorisation models, assessment of data recognition quality, and detection of anomalies. The research methodology consisted of the following steps: formulation of the problem, selection of data and labelling, construction of the model for prediction, installation, and testing. The data tagging technique was based on the requirements set by the Global Aviation Organisation for business jet-engine aircraft, which expert business pilots then validated. The 93% precision demonstrated an excellent match for the most effective prediction model, linear dipole testing. Furthermore, the "good fit" of the model was verified by its achieved area-under-the-curve ratios of 0.97 for abnormal identification and 0.96 for daily detection.