Open AccessAn Efficient Ancient Tamil Script Classification System using Gradient Boosted Tree Algorithm
Author(s) -
T.S. Suganya*,
S. Murugavalli
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.c4943.098319
Subject(s) - tamil , variety (cybernetics) , computer science , hough transform , artificial intelligence , conch , boosting (machine learning) , process (computing) , decision tree , feature (linguistics) , tree (set theory) , natural language processing , pattern recognition (psychology) , mathematics , geology , linguistics , literature , art , image (mathematics) , paleontology , programming language , combinatorics , philosophy
A variety of historical information’s has been derived and obtained from the various forms of inscriptions existing worldwide. Regional languages of certain selected sources have been employed in the inscription writing procedures. One such regional language with a wealthy literature and heritage that can be suitably opted for the inscription writing process is ‘Tamil’. Encryption of the writing was done with the help of certain materials such as metals, conch shells, stones, palm leaf, and copper plate. These are found to be rich in the data contents related to the fields of history, religious, administrative, astronomy, culture, economic tax and educational conditions. This paper incorporates certain feature extraction methods like Shape and Hough transform and Gradient Boosting tree as classifiers for identifying the ancient Tamil script.