METHOD FOR AUTOMATED ANALYSIS OF REFERENCE FORMS Russian patent published in 2016 - IPC G06F17/21 G06F21/60 

FIELD: information technology.

SUBSTANCE: invention relates to means for automated analysis of text documents. All information significant fields of reference document forms are converted with indication of their positions in each form. Fields of reference document forms are detected in electronic form of analysed document. Number of fields of reference document forms are detected in analysed document. Reference forms are searched in analysed document. A list of reference document forms found in analysed document is created. Order of fields in each reference document form in analysed document is checked, and is deleted if said order does not match in a specific reference form. Position of fields of reference document forms is determined from displayed list in analysed document. Text is identified in analysed document, found between fields found. Availability of said reference document form in filled form in analysed text is determined, if text is detected between any adjacent fields. Name of current form is placed in list of reference document forms found in analysed text.

EFFECT: technical result is improved accuracy of determining presence of confidential information in documents.

3 cl, 1 dwg