A study of methods of classification learning with determination coefficients separating surfaces by using optimization methods in the base of wavelet transform with and without constraints (in the form of inequalities) is carried out. Recommendations for selecting optimization method for the classification in the automated systems are allowed for the stage of the production process is propozed. These recommendations can justify the choice of optimization method for the classification depending on the stage of training (additional training) of the automated system and, as a rule, conflicting demands on speed and accuracy of classification at the specific situation.
automated systems, classification, optimization, wavelet transform
“Issledovanie avtomatizirovannoi klassifikatsii dannykh izmerenii s ispolzovaniem veivlet-preobrazovaniia” [Research of the automated classification using the wavelet transform],
Information Processing Systems,