Extracting structure from text documents based on machine learning

This study is devoted to a method that facilitates the task of extracting structure from text documents using an artificial neural network. The consists data preparation, building and training model results evaluation. Data preparation includes collecting corpora documents, converting variety file formats into plain text, manual labeling each document structure. Then are split tokens paragraphs. paragraphs represented as feature vectors provide input trained validated on selected subsets. Trained evaluation presented. final performance calculated per label precision, recall, F1 measures, overall average. can be used extract sections bearing similar