Character Recognition of Handwriting of Javanese Character Image using
Information Gain Based on the Comparison of Classification Method
Dublin Core
Title
Character Recognition of Handwriting of Javanese Character Image using
Information Gain Based on the Comparison of Classification Method
Information Gain Based on the Comparison of Classification Method
Subject
character recognition; javanese character; information gain; LBP
Description
ndonesia is a country rich in a variety of regional cultures. Regional airspace needs to be preserved so as not to become
extinct. One of them is the local culture of Central Java Province, namely Javanese Character. In this modern era, globalization
is growing in every country. The impact of globalization is increasingly widespread and developing in society. One effect of
globalization is local people prefer foreign language skills to learn local languages. This study, appliesthe method of character
recognition using a new combination workflow that contains Local Binary Pattern (LBP) and Information Gain. Then compare
Support Vector Machine (SVM), k-Nearest Neighbor and Naïve Bayes. The LBP method is used to obtain an image's texture
or shape characteristics. Information Gain is used for the feature selection algorithm, whereas SVM, k-Nearest Neighbor and
Naïve ayes is used for the classification method. From previous research, the information gain method succeeded in increasing
the accuracy by 2%. This research compares the SVM classification with another classification method, and the result shows
that our proposed can improve classification performance. The best accuracy result using SVM classification gets 87,86%, at
ten folds and cell size 64x64.
extinct. One of them is the local culture of Central Java Province, namely Javanese Character. In this modern era, globalization
is growing in every country. The impact of globalization is increasingly widespread and developing in society. One effect of
globalization is local people prefer foreign language skills to learn local languages. This study, appliesthe method of character
recognition using a new combination workflow that contains Local Binary Pattern (LBP) and Information Gain. Then compare
Support Vector Machine (SVM), k-Nearest Neighbor and Naïve Bayes. The LBP method is used to obtain an image's texture
or shape characteristics. Information Gain is used for the feature selection algorithm, whereas SVM, k-Nearest Neighbor and
Naïve ayes is used for the classification method. From previous research, the information gain method succeeded in increasing
the accuracy by 2%. This research compares the SVM classification with another classification method, and the result shows
that our proposed can improve classification performance. The best accuracy result using SVM classification gets 87,86%, at
ten folds and cell size 64x64.
Creator
Irham Ferdiansyah Katili1
, Mochamad Arief Soeleman2
, Ricardus Anggi Pramunendar3
, Mochamad Arief Soeleman2
, Ricardus Anggi Pramunendar3
Publisher
Universitas Dian Nuswantoro Semarang
Date
: 06-02-2023
Contributor
Fajar bagus W
Format
PDF
Language
Indonesia
Type
Text
Files
Collection
Citation
Irham Ferdiansyah Katili1
, Mochamad Arief Soeleman2
, Ricardus Anggi Pramunendar3, “Character Recognition of Handwriting of Javanese Character Image using
Information Gain Based on the Comparison of Classification Method,” Repository Horizon University Indonesia, accessed June 6, 2025, https://repository.horizon.ac.id/items/show/9330.
Information Gain Based on the Comparison of Classification Method,” Repository Horizon University Indonesia, accessed June 6, 2025, https://repository.horizon.ac.id/items/show/9330.