Recognition and grouping of digital contents based on semantic

U ovom radu analizirani se modeli temeljeni na statistikama i modeli temeljeni na strojnom učenju koji su pogodni za obradu tekstualnih digitalnih sadržaja. Kroz rad se obradila obrada prirodnog jezika, automatizirano prolanaženje pojmova u jednostavnom tekstu, izrada grafa semantičkih odnosa, semantička sličnost dvaju tekstova, te rad u alatu Tropes.In this work models based on statistics and models based on machine learning that are suitable for processing text digital content were processed. Through the work it has also been processed natural language processing, automatic concept extraction from plain text, semantic relationship graph, semantic similarity of the two texts, and work in the tool Tropes