NTRS - NASA Technical Reports Server

Back to Results
An Artificial Intelligence Classification Tool and Its Application to Gamma-Ray BurstsDespite being the most energetic phenomenon in the known universe, the astrophysics of gamma-ray bursts (GRBs) has still proven difficult to understand. It has only been within the past five years that the GRB distance scale has been firmly established, on the basis of a few dozen bursts with x-ray, optical, and radio afterglows. The afterglows indicate source redshifts of z=1 to z=5, total energy outputs of roughly 10(exp 52) ergs, and energy confined to the far x-ray to near gamma-ray regime of the electromagnetic spectrum. The multi-wavelength afterglow observations have thus far provided more insight on the nature of the GRB mechanism than the GRB observations; far more papers have been written about the few observed gamma-ray burst afterglows in the past few years than about the thousands of detected gamma-ray bursts. One reason the GRB central engine is still so poorly understood is that GRBs have complex, overlapping characteristics that do not appear to be produced by one homogeneous process. At least two subclasses have been found on the basis of duration, spectral hardness, and fluence (time integrated flux); Class 1 bursts are softer, longer, and brighter than Class 2 bursts (with two second durations indicating a rough division). A third GRB subclass, overlapping the other two, has been identified using statistical clustering techniques; Class 3 bursts are intermediate between Class 1 and Class 2 bursts in brightness and duration, but are softer than Class 1 bursts. We are developing a tool to aid scientists in the study of GRB properties. In the process of developing this tool, we are building a large gamma-ray burst classification database. We are also scientifically analyzing some GRB data as we develop the tool. Tool development thus proceeds in tandem with the dataset for which it is being designed. The tool invokes a modified KDD (Knowledge Discovery in Databases) process, which is described as follows.
Document ID
Document Type
Hakkila, Jon (Charleston Coll. Charleston, SC, United States)
Haglin, David J. (Minnesota State Univ. Mankato, MN, United States)
Roiger, Richard J. (Minnesota State Univ. Mankato, MN, United States)
Giblin, Timothy (Charleston Coll. Charleston, SC, United States)
Paciesas, William S. (Alabama Univ. Huntsville, AL, United States)
Pendleton, Geoffrey N. (Alabama Univ. Huntsville, AL, United States)
Mallozzi, Robert S. (Alabama Univ. Huntsville, AL, United States)
Date Acquired
August 21, 2013
Publication Date
January 1, 2004
Subject Category
Funding Number(s)
Distribution Limits
Work of the US Gov. Public Use Permitted.
Document Inquiry