Development of a method for determining the indicators of manipulation based on morphological synthesis

Research on the development of methods for identifying signs hidden manipulation (destructive information and psychological impact) in text messages that are published Internet sites distributed among users social networks is relevant. One main problems these difficulty formalizing process network agents. To do this, based morphological synthesis, it necessary to determine relevant indicators analyzing criteria making a decision about presence messages. Based method determining was developed, taking into account achievements modern technologies intelligent content analysis messages, machine learning methods, fuzzy logic computational linguistics, which made possible reasonably group evaluating manipulation. The stages include message at level perception by indicator readability, phonetic emotional impact subconscious, graphic marking intensity, calculating integral message. proposed method, specialized software developed provided 13 % greater accuracy manipulative compared known expert evaluations, reduced influence subjective factor evaluation result