Olga Tabolina

  • Citations Per Year
Learn More
The problem of alignment of two symbol sequences is considered. The validity of the available algorithms for constructing optimal alignment depends on the weighting coefficients which are frequently difficult to choose. A new approach to the problem is proposed, which is based on the use of vector weighting functions (instead of tradionally used scalar(More)
This article presents a technique, namely, structured neural network, to model the network of cause-and-effect relationships of the loyalty-profit chain for a chemical industry. A comparison between the structured neural network, the traditional neural network and regression models is presented. It is concluded that a strictly empirical modeling approach is(More)
  • 1