R Sulej et al 2007 Meas. Sci. Technol. 18 2486 doi:10.1088/0957-0233/18/8/025
R Sulej1, K Zaremba1, K Kurek2 and E Rondio2
Show affiliationsIn this paper, we present the application of a neural network for events classification in a high-energy physics experiment. As a network model we use a multi-layer perceptron with a dynamic topology adjustment algorithm. Our solution covers both adding new hidden neuron units and removing unnecessary units. Neural network results are compared to the standard kinematical cuts techniq1guuuuue and to the well-known k-nearest neighbour classifier.
07.05.Mh Neural networks, fuzzy logic, artificial intelligence
Accelerators, beams and electromagnetism
Issue 8 (August 2007)
Received 31 October 2006, in final form 10 January 2007
Published 6 July 2007
R Sulej et al 2007 Meas. Sci. Technol. 18 2486
Patrick L T M Frederix et al 2005 Nanotechnology 16 997
O Gzowski et al 1981 J. Phys. D: Appl. Phys. 14 L77
Ronald W Hellings 2003 Class. Quantum Grav. 20 1019
C Champion and C Le Loirec 2006 Phys. Med. Biol. 51 1707
Pål Erik Goa et al 2001 Supercond. Sci. Technol. 14 729
Tamara Bechtold et al 2005 J. Micromech. Microeng. 15 1205
J Mitroy and M W J Bromley 2003 J. Phys. B: At. Mol. Opt. Phys. 36 793
D J Baker et al 2007 Phys. Scr. 75 615
Claus Lämmerzahl 1998 Class. Quantum Grav. 15 13