National Technical University of Athens (NTUA)
Μοντελοποίηση της ελληνικής νοηματικής γλώσσας για τα συστήματα στατιστικής μηχανικής μετάφρασης
Abstract
dc:descriptionThis thesis is located in the framework of Automatic Machine Translation and in the human and machine software interface for hearing disabled people using the Greek Sign Language. In this work we present a novel prototype Rule Based Machine Translation (RBMT) system for the creation of large quality written Greek Sign Language (GSL) glossed corpora. In particular, the proposed RBMT system supports the professional translator of GSL to produce different kinds of GSL glossed corpus. Then the glossed corpus is used as training data for the production/creation of Language Model (LM) n-gram. With the GSL glossed corpus and for any domain, we can build, test and evaluate different kinds of Language Models for different kinds of glossed GSL corpus, even if there is no real written GSL large corpus. These GSL parallel corpus and languages models also will be used as training data by the Statistical Machine Translation (SMT) MOSES application system. It should be noted that the whole process is robust and flexible, since it does not demand deep grammar knowledge of GSL. By using the BiLingual Evaluation Understudy (BLEU) metric score, our prototyped MT system achieves a very promising performance, and in particular an average score of 60,53% και 85,1% / 65,5% / 53,8% / 44,8% για 1-gram / 2 -gram / 3-gram / 4-gram.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2020
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kouremenos, Dimitrios
- Κουρεμένος, Δημήτριος
Subjects
dc:subject × 16- Στατιστική μηχανική μετάφραση
- Ελληνική γλώσσα
- Ελληνική νοηματική γλώσσα (ΕΝΓ)
- Επικοινωνία κωφού - μηχανής
- MOSES
- Μοντέλο φράσης
- Statistical machine translation
- Greek language
- Greek sign language (GSL)
- Deaf people communication
- Phrase model
- gr
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Natural Sciences
- Computer and Information Sciences
Rights
dc:rights- Statement dc:rights
-
- BY
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/48061