The article is devoted to the development of a method of clustering of text documents based on vector representation of text corpus using the graph model. The proposed method allows the clustering of text documents, taking into account the weighting factors of individual words in the corpus. The proposed method uses a cosine measure as the distance between the documents and can be used for the structuring text information corps of large dimension.
text document, vectorial presentation, count model, cosine distance
"Metod vektorno-hrafovoi klasteryzatsyy dokumentov v systemakh obrabotky tekstovoi ynformatsyy" [The vector-graph’s clustering method of documents in text processing systems],
Information Processing Systems,