NASA Logo, National Aeronautics and Space Administration

+NASA Home

+Ames Home

Data Sciences Group Presents Work at Machine Learning and Data Analytics Symposium in Doha, Qatar
Intelligent Systems Division Banner

Data Sciences Group Presents Work at Machine Learning and Data Analytics Symposium in Doha, Qatar

Bryan Matthews gave an invited talk at the fourth annual Machine Learning and Data Analytics (MLDAS) Symposium on March 14th, 2017, in Doha, Qatar. The talk was entitled, “Applications of Anomaly Detection and Precursor Identification in Airspace Operations” and discusses innovative data mining work being done in the Data Sciences Group within the Airspace Operations and Safety Program. The Symposium was highlighted in the local media outlet, Gulf Times.

BACKGROUND: The Data Sciences Group has been working to further aviation safety by developing algorithms that can provide anomaly detection and precursor identification capabilities while exploring Big Data problems. These methods analyze data from various airspace operations and automatically identify operationally significant anomalous behavior within the national airspace. This work has two main areas of focus:

  1. The anomaly detection algorithm has the ability to jointly analyze heterogeneous data types using multiple kernel techniques and one-class support vector machines. This approach combines diverse data to identify anomalies that would not otherwise be detected from a single homogeneous source. With this capability, safety analysts can discover new anomalies that were previously overlooked and better assess new safety risks in the airspace.
  2. The precursor identification algorithm utilizes both inverse reinforcement and reinforcement learning to build a model that can identify suboptimal actions in time series data. When an action that leads to unrecoverable sequences of events results in an adverse event, a potential precursor is flagged. This tool has the potential to be utilized in near real time to help identify periods in a flight where the action taken may lead to an adverse event.
The MLDAS Symposium was jointly sponsored by Boeing and the Qatar Computing Research Institute. The purpose of the Symposium was to bring together researchers, practitioners, students, and industry experts in the fields of machine learning, data mining, and related areas to present recent advances, discuss open research questions, and help bridge the gap between data analytics research and industry needs on certain concrete problems. In particular, the MLDAS Symposium aimed to serve as a platform for the exchange of ideas, identification of important and challenging applications, and discovery of possible synergies.

NASA PROGRAM FUNDING: Shadow Mode Assessment using Realistic Technologies for the National Airspace System (SMART-NAS) project, Airspace Operations and Safety Program (AOSP), NASA Aeronautics Mission Directorate (ARMD)

COLLABORATORS: TI: Vijay Janakiraman (USRA), Bryan Matthews (SGT), David Nielsen (Mori), and Nikunj Oza; ATAC CORP: Kennis Chan and John Schade

POINT OF CONTACT: Nikunj Oza, nikunj.c.oza@nasa.gov

First Gov logo
NASA Logo - nasa.gov