Modeling of Process Dynamics by Sequence of Homogenous Semantic Networks on the Base of Text Corpus Sequence Analysis
The paper describes an approach to modeling of process dynamics based on the technology of automatic semantic analysis of textual information. During processing of text an associative network is generated, the key concepts of which, including lexical markers of the analyzed process, are ranked by their semantic weight. Defined by the marker status on the “good-bad” scale, this weight provides the marker value for characterization of the process state. It is changes in the normalized characteristic of the process combined for all markers from time sample to time sample that characterize the process trend.