Application of the Hidden Markov Model for Innovative Projects "Viability" Analysis

This master thesis deals with determining of innovative projects "viability". "Viability" is the probability of innovative project being implemented. Hidden Markov Models are used for evaluation of this factor. The problem of determining parameters of model, which produce given data sequence with the highest probability, are solving in this research. Data about innovative projects contained in reports of Russian programs "UMNIK", "START" and additional data obtained during study are used as input data for determining of model parameters. The Baum-Welch algorithm which is one implementation of expectation-maximization algorithm is used at this research for calculating model parameters. At the end part of the master thesis mathematical basics for practical implementation are given (in particular mathematical description of the algorithm and implementation methods for Markov models).

Search of Method for Analyzing "Viability" of Innovative Projects

Questions of "viability" evaluation of innovation projects are considered in this article. As a method of evaluation Hidden Markov Models are used. Problem of determining model parameters, which reproduce test data with highest accuracy are solving. For training the model statistical data on the implementation of innovative projects are used. Baum-Welch algorithm is used as a training algorithm.

Dynamic potential fields for guided exploration in virtual environments

Dynamic potential fields, guided exploration, virtual environments, navigation

Instability of excitation waves induced by electrical fields

Pyrogallol, uncatalyzed bromate oscillator, electric field, pulse wave, Belousov-Zhabotinsky reaction, reversal

Extraction of topological structures in 2D and 3D vector fields

feature extraction, feature tracking, vector field visualization

Extraction of vortex structures in 3D flow fields

Magdeburg, Univ., Fak. für Informatik, Diss., 2009

Conversive Hidden non-Markovian models

Magdeburg, Univ., Fak. für Informatik, Diss., 2012

Lagrangian methods for visualization and analysis of time-dependent vector fields

Magdeburg, Univ., Fak. für Informatik, Diss., 2013

Shapes in vector fields : methods for continuous deformations and surface-based flow visualizations

Magdeburg, Univ., Fak. für Informatik, Diss., 2013

Optimal design in the presence of random or fixed block effects

Magdeburg, Univ., Fak. für Mathematik, Diss., 2014

Approximations and asymptotic expansions for sums of heavy-tailed random variables

Magdeburg, Univ., Fak. für Mathematik, Diss., 2015

Optimal designs for the prediction in hierarchical random coefficient regression models

Magdeburg, Univ., Fak. für Mathematik, Diss., 2015

Topology of vector fields with uncertainty

Magdeburg, Univ., Fak. für Informatik, Diss., 2015