Correlation Construction and Analysis of Segment Network Based on Improved FP-Growth Algorithm

Addressing the challenge of identifying pivotal segments within the aviation network, exacer-bated by the escalating air traffic and airspace congestion, this study employs the FP-Growth algorithm to enhance the correlation analysis of the segment network's edges. This enhancement facilitates the construction and detailed examination of the segment network's correlation model. The methodology commences with the development of a segment correlation network model, leveraging relative hourly flight traffic data. Subsequently, a thorough analysis of the model network's topological characteristics is conducted. The findings reveal that the refined FP-Growth algorithm outperforms both the standard FP-Growth and Apriori algorithms in terms of efficiency and accuracy. Moreover, the application of degree centrality, pagerank, and eigenvector centrality successfully identifies influential segments, such as ENH-P373 and YIH-ENH, which play a crucial role in the operational efficiency and overall safety of the aviation network. This research provides a solid theoretical foundation for the precise identification and management of critical network segments.