Open AccessA method for comparing text and graphic fragments in electronic documents using a hybrid criterion.
Author(s) -
Serhii Udovenko,
Larysa Chala,
Є. С. Кушвід
Publication year - 2019
Publication title -
bionika intellekta
Language(s) - English
Resource type - Journals
eISSN - 2663-306X
pISSN - 2663-3051
DOI - 10.30837/bi.2019.1(92).12
Subject(s) - computer science , information retrieval , similarity (geometry) , image (mathematics) , value (mathematics) , data mining , artificial intelligence , machine learning
A method for comparing text and graphic fragments in electronic documents by a hybrid criterion is considered. This method allows you to determine the integrated similarity value between the request associated with the image in the request and the text-graphic image in the document database. The criterion used provides for the use of weights for images with annotations and for images without annotations. The prospects for using the proposed method are determined