| dc.description.abstract |
Today information technologies play an increasingly significant role in society. The technology development has already reached such a level that large corporations are beginning to use systems not only for processing information, but also for forecasting certain processes in society, nature, economy and other spheres of human life. One of these ways of processing information without using human resources is neural networks. The scientific novelty is determined by the fact that the article explores the symbolic analysis of text using a neural network. Recognition of written characters using a neural network could be useful for psychological studies of the relationship between a person's psychological state and his handwriting. As can be found out in a person's handwriting, the size of a slope, the direction of the letters, the pressure on a pen and the nature of lettering itself are important. If to combine a neural network for recognising written signs and a database of handwriting of many people, interesting statistics can be seen and the attempt to find connection between handwriting and character can be made. The practical significance of the research is determined by the fact that the research results can be used by specialists in different fields. The main problem in these cases is an optimised written character recognition algorithm that can correctly and quickly recognise written characters. Based on this, the object of research is the algorithm for recognising written characters, and the subject of research is the neural network itself. | en_US |
