Malaysian to German sign language statistical machine translation using Markov chain and search algorithms

Closed

Fridy Mandita, Harnan Malik Abdullah, Toni Anwar

2017 Proceedings of the 2017 4th International Conference on Computer Applications and Information Processing Technology, CAIPT 2017 Vol. 2018-January Conference paper Cited by 0 Quartile

Abstract

Sign language has shown advancement indication results in the last decade of years. Research in the area of sign language has been done but mostly focuses in signer independent schemas consists of sign language and video. Unfortunately, research of sign language to translate one sign language to another sign language is infrequently. The research contributes to Malaysian to German sign language statistical machine translation for hearing impaired people. Markov chain and search algorithms are used to apply the process of translation from Malaysian to German sign language and vice versa. The corpus data are used for testing statistical machine translation around 10.000. A binary search and linear search algorithm in cooperation with a Markov chain are used to apply the process of translation. An output of execution time between 2 models of algorithms has been explored. The accuracy of the binary search algorithm is better than linear search algorithm when using for translation. © 2017 IEEE.

Affiliations

Informatics Department, University of 17 Agustus 1945, Surabaya, Indonesia; Diploma in Computer Engineering, University of Brawijaya, Malang, Indonesia; Department of Software Engineering, Universiti Teknologi Malaysia Johor, Bahru, Malaysia