Author Identification - An Experiment Based on Mongolian Literature Using Decision Trees

Author

Damiran, Zolboo ; Altangerel, Khuder

Author_Institution

Dept. of Inf. Syst., MUST-CSMS, Ulaanbaatar, Mongolia

fYear

2014

fDate

12-14 July 2014

Firstpage

186

Lastpage

189

Abstract

Nowadays, cases of piracies, copyrights, legal disputes and allegations are increasing as far as software is concerned. In such cases it is difficult to say who is right and who is wrong between the two quarreling parties due to insufficient evidences. So, what can be done. Author identification plays a very important role in most of these cases, such as plagiarism detection, masquerade detection, software maintainability and resolving authorship disputes. In this research, we have analyzed the result of an experiment on a text classification using Decision tree method approaches for Mongolian language. We have a training set of documents, each labeled with one of 13 classes. Since it is the first work in this field for Mongolian language no previous work results were available for comparison.

Keywords

computer crime; copyright; decision trees; natural language processing; pattern classification; software maintenance; text analysis; Mongolian literature; allegations; author identification; authorship disputes; copyrights; decision tree method; document training set; legal disputes; masquerade detection; piracies; plagiarism detection; software maintainability; text classification; Accuracy; Classification algorithms; Decision trees; Entropy; Software; Text categorization; Training; Decision Trees method; Machine learning; Text classification;

fLanguage

English

Publisher

ieee

Conference_Titel

Ubi-Media Computing and Workshops (UMEDIA), 2014 7th International Conference on

Conference_Location

Ulaanbaatar

Print_ISBN

978-1-4799-4267-1

Type

conf

DOI

10.1109/U-MEDIA.2014.22

Filename

6916349