Другие журналы
|
Stroganov
Formalized curriculum model for optimization of individual learning paths
Engineering Education # 11, November 2012 DOI: 10.7463/1112.0506173 The authors solve the problem of constructing a formal model of curriculum and structuring requirements for the minimum specialist training. They propose keeping a terminological glossary that provides the basis for determining performance criteria of developed curricula. Algorithms for calculating the function of forgetting terms are given in the article. The authors consider evaluation of connectivity of disciplines and optimization of the sequence of modules in order to maximize the values of learning functions in all terms. Operations on disciplines, modules and terms for implementation of search operations and editing base operations are listed.
Evaluation of efficiency of adaptive test control procedures
Engineering Education # 11, November 2012 DOI: 10.7463/1112.0506146 The authors consider formal description of procedures for computer control of knowledge. For the adaptive algorithm of presentation of tests a model of evaluation of its effectiveness in minimizing the probability of misclassification was created. To solve this problem the authors propose to use the Markov chains whose states correspond to the levels of complexity, and transition probabilities are calculated basing on Birnbaum's logistic model. The ratio for stationary probabilities which show a significant advantage of adaptive algorithms for test control were obtained.
77-30569/359132 Influence of initial conditions and time of simulation on characteristics of conditional non-stationary processes
Engineering Education # 04, April 2012 In the article a provisionally non-stationary Gaussian process was proposed for analytical research of simulation processes convergence. This Gaussian process was considered as a stationary Gaussian process with a specified pre-history. The authors studied dependence of medium-integral estimation on different factors. It was shown that the trend of the medium-integral estimation had a lasting character, contains a significant systematic error at the initial stage and depends on the process correlation, initial values and time of simulation.
77-30569/355371 Approximation of integral mean estimations of transient regimes of simulated models of queuing networks
Engineering Education # 03, March 2012 This article deals with the problems of estimation of the character of autocovariance functions and trends of typical output simulated processes, which are necessary when investigating the convergence of search optimization algorithms on simulated models. In order to determine the function class of output simulated processes, simulation of typical structures of queuing networks was conducted, namely disconnected single- and multi-phase queue systems with different service procedures and closed-loop single- and multi-phase queue systems with sources of various population and service procedures.
|
|
|||||||||||||||||||||||||||||
|