Open AccessAPMS SVD methodology and implementation
Author(s) -
BG Amidan,
TA Ferryman
Publication year - 2000
Language(s) - English
Resource type - Reports
DOI - 10.2172/753847
Subject(s) - computer science , singular value decomposition , graphics , aviation , decomposition , value (mathematics) , data mining , artificial intelligence , machine learning , computer graphics (images) , engineering , aerospace engineering , ecology , biology
One of the main tasks within the Aviation Performance Measurement System (APMS) program uses statistical methodologies to find atypical flights. With thousands of flights a day and hundreds of parameters being recorded every second for each flight, the amount of data escalates and the ability to find atypical flights becomes more difficult. The purpose of this paper is to explain the method known as single value decomposition (SVD) employed to search for the atypical flights and display useful graphics that facilitate understanding the causes of atypicality for these flights. Other methods could also perform this search and some are planned for future implementation