The methods of classifying handwriting samples as to their type - Classifiers

Published: 1 March 2022| Version 1 | DOI: 10.17632/ng5x5jfss8.1
Contributor:
Jagoda Dzida

Description

The discriminant functions of Classifiers - the methods created to classify handwriting samples as to their type. The data are linked to the research article.

Files

Steps to reproduce

The Classifiers were created using the discriminant analysis of the Statistica 13.1 computer programme. The Classifiers consist of the discriminant functions for every type of handwriting sample. If we take into consideration 11 possible types of handwriting samples, we have 11 discriminant functions. The methods calculate the results for every discriminant function and classify the sample by choosing the discriminant function with the highest result. For example, if the highest result was calculated for the discriminant function of a natural sample, our handwriting sample is probably a natural one.

Institutions

Uniwersytet im Adama Mickiewicza w Poznaniu

Categories

Document Analysis, Handwriting Questioned Document

Licence