详情

原文传递 Natural Language Understanding and Extraction of Flight Constraints Recorded in Letters of Agreement.

题名：	Natural Language Understanding and Extraction of Flight Constraints Recorded in Letters of Agreement.
作者：	Clarke, S. S. B; Zhu, Z; He, O; Almeida, J. A. A; Kalyanam, K; Pai, R.
摘要：	This paper presents an automated information extraction and inference technique usingnatural language processing for extracting flight operational procedures and constraintsembedded in heritage air traffic management documents. The extracted flight constraints canbe digitized and fit into existing airspace information exchange models such as theAeronautical Information Exchange Model (AIXM). This approach offers a digitized solutionto disseminate airspace operating conditions to diverse air users and stakeholders in theNational Airspace System (NAS). Furthermore, the digitized flight procedures can provideoperational flexibility for emerging advanced air mobility providers and reduce trafficcontroller workload while maintaining current safety standards. To demonstrate this process,1,972 Letters of Agreement (LOAs) have been selected for processing, named entityextraction, constraint identification and extraction. This dataset is derived from a subset ofdocuments related to Air Route Traffic Control Centers (ARTCC) operations. Weexperimented with various traditional information extraction techniques, state-of-the-artmachine learning and deep learning models to perform named entity recognition and patternrecognition on our dataset. We present the results from our experiments and demonstrate99.0% F-1 score for named entity recognition, and a 96.6% accuracy for our entire workflowup to named entity recognition. We also discuss constraint definitions using generic patternedtemplates and extensions to this work in applying entity linking to digitally extracting relevantconstraints.
总页数：	16 pages