articles

The article proposes a new approach to intervention analysis on the economic performance of the organization using the neural network of problem oriented architecture. The results of model application experiments using the Yota company turnover data demonstrated better standard deviation values versus the corresponding values for ARIMA based model.

The models for comparative analysis of innovation based on the use of information approach A. A. Denisov are proposed. The models may be used for evaluating the innovation effects regarding their impact on the achievement of the organization’s goals and feasability issues as well as for analyzing innovations development over time and taking into account their mutual influence.

Architectural models that become popular in the last 10–15 years are especially important at the meso- and macro-levels. The article reveals the concept of the so-called e-government architecture. In this regard progress of leading countries such as the U. S., Britain, Germany and Russia in this area has been examined and the architecture of a hypothetical government organizational design as well as the methodology for its projection in the documents were defined.

This article presents various techniques for building fast recommender systems based on dimension reduction and classification of web site usage data. The main web site types using recommendation systems are given and their features discussed.

The problem of compression of hyperspectral remote sensing data is considered and a multi stage algorithm proposed. Formation of a large redundancy auxiliary data based on their byte representation and accounting inter-channel correlation is used to increase the compression ratio of the algorithm. Some results of studies on the effectiveness of compression of hyper spectral space images made by the proposed compression algorithm with the purpose and specialized archiving software are given.

A MLP based algorithm for combining expert opinions using algorithmic composition with a dynamic mechanism to assess the coefficient of confidence in each expert is suggested. In order to increase the effectiveness of the experts the possibility of modifying the standard gradient algorithms learning methods based on the use of combined heuristics as well using experts in the Intrusion Detection System in the network is discussed.