The Cascade Orthogonal Neural Network

In the paper new non-conventional growing neural network is proposed. It coincides with the Cascade- Correlation Learning Architecture structurally, but uses ortho-neurons as basic structure units, which can be adjusted using linear tuning procedures. As compared with conventional approximating neural networks proposed approach allows significantly to reduce time required for weight coefficients adjustment and the training dataset size.

Оптимизация на учителския състав в училище

Silvia Baeva - In the Ministry of Education and Science’s system it has been talked about optimization of the school network; this optimization can be carried out in different directions and be supported by laws. An important aspect in the optimization of the school network is to reduce costs and increase overall efficiency in each school. We will formulate this aspect as a problem of the multicriteria decisions making and by appropriate numbers of methods and criteria it can be transformed to the problem of unicriterion optimization. This problem is separated into two stages: 1-th stage – determining the minimum number of classes in a school under certain statutory provisions for the size of each of them; 2-th stage – the appointment of a minimum number of teachers and achievement of maximal effectiveness of teaching and acquiring of knowledge and skills by students according to certain statutory provisions for teachers’ annual norms of subjects and number of hours in a particular subject area and the salaries of teachers. The achievement of maximum overall efficiency is a priority in all schools.