Item


Tsallis Mutual Information for Document Classification

Mutual information is one of the mostly used measures for evaluating image similarity. In this paper, we investigate the application of three different Tsallis-based generalizations of mutual information to analyze the similarity between scanned documents. These three generalizations derive from the Kullback鈥揕eibler distance, the difference between entropy and conditional entropy, and the Jensen鈥揟sallis divergence, respectively. In addition, the ratio between these measures and the Tsallis joint entropy is analyzed. The performance of all these measures is studied for different entropic indexes in the context of document classification and registration

This work has been funded in part with Grant Numbers TIN2010-21089-C03-01 from the Spanish Government and 2009-SGR-643 from the Catalan Government.

MDPI

Author: Vila Duran, Marius
Bardera i Reig, Antoni
Feixas Feixas, Miquel
Sbert, Mateu
Abstract: Mutual information is one of the mostly used measures for evaluating image similarity. In this paper, we investigate the application of three different Tsallis-based generalizations of mutual information to analyze the similarity between scanned documents. These three generalizations derive from the Kullback鈥揕eibler distance, the difference between entropy and conditional entropy, and the Jensen鈥揟sallis divergence, respectively. In addition, the ratio between these measures and the Tsallis joint entropy is analyzed. The performance of all these measures is studied for different entropic indexes in the context of document classification and registration
This work has been funded in part with Grant Numbers TIN2010-21089-C03-01 from the Spanish Government and 2009-SGR-643 from the Catalan Government.
Document access: http://hdl.handle.net/2072/253808
Language: eng
Publisher: MDPI
Rights: Attribution 3.0 Spain
Rights URI: http://creativecommons.org/licenses/by/3.0/es/
Subject: Informaci贸, Teoria de la
Information theory
Classificaci贸
Classification
Title: Tsallis Mutual Information for Document Classification
Type: info:eu-repo/semantics/article
Repository: Recercat

Subjects

Authors