You are on page 1of 17

ASSOSA UNIVERSITY

COLLEGE OF COMPUTING AND INFORMATICS


DEPARTMENT OF INFORMATION TECHNOLOGY

A Thesis Proposal
on
English to Amharic Neural Machine Translation

By Kirubel Andargie
Advisor :- Yimer A(Asst. Prof) July 2022
Assosa, Ethiopia
Outlines
 Introduction

 Motivation

 Statement of Problem

 Objectives of the Study

 Scope and limitation of Study

 Methodology

English to Amharic Neural Machine


Translation
Introduction
 Natural language is one of the fundamental aspects of human behavior

and is a crucial component in our lives

 It is the method of communication in different ways which are by audio

(spoken), text (written) and sign to exchange ideas, emotions, and

information

 The advancement of technology and the rise of the internet led to an

ever-increasing demand for Natural Language Processing

English to Amharic Neural Machine


Translation
Introduction Cont’d

 NLP applications are useful in facilitating human-human, human-

computer, computer-human, and computer-computer communication via

computing systems

 In the Natural Language Processing world, each and every language should

be well understood by the machine so as to communicate very well

 The process that lets the machine understand the different languages used

all around the world is called machine translation


English to Amharic Neural Machine
Translation
Introduction Cont’d
 Machine Translation (MT) is a process by which
computer software is used to translate a text from
one natural language to another

 Translation is not a mere word-for-word substitution

 A translator must interpret and analyze all of the


elements in the text and know how each word may
influence another
English to Amharic Neural Machine
Translation
Introduction Cont’d
A number of approaches are available for MT. Such as –
 Rule-Based Machine Translation (RBMT),
 Corpus-based Machine Translation Approach (Corpus
based MT),
 Hybrid Machine Translation (Hybrid MT) and
 Neural Machine Translation (NMT)
 NMT is a recently proposed framework for machine
translation based purely on neural networks.
English to Amharic Neural Machine
Translation
Motivation
 Even though Ethiopia is the oldest independent nation in Africa , the
country has not been using the technology as well as it should
 Despite of the fact that almost all devices use English language, there
must be a way to facilitate the implementation of every application
 Most important data such as literatures, e-books, research documents,
papers and other raw facts are written in English

 As a developing country, Ethiopia needs this information, Since it is


inconvenient to document all the data in hard copies by translating them
to Amharic, people need a way to access the data only when necessary

English to Amharic Neural Machine


Translation
Statement of Problem
 People use human translation and they tend to be slower as

compared to machines

 In addition, it can be more important to get the result without

delay which is hard to accomplish with a human translator

 Though a skilled human translator will deliver a better quality

translation, there are many times when machine translation is

sufficient for the task, which will help save you time and money
English to Amharic Neural Machine
Translation
Statement of Problem Cont’d
 Several studies have been done on language pair of English and

other languages, such as for

 Arabic-to-English Neural Machine Translation and

 English-Japanese Machine Translation,

 French to English Statistical Machine Translation system

 However, only little work has been done on the MT system

among English and Ethiopian languages


English to Amharic Neural Machine
Translation
Statement of Problem Cont’d

 Some of the studies are carried out on

 English-Amharic language pair using SMT,

 English-Afaan Oromo language pair using SMT

 Some of the MTs done between Ethiopian languages are

 Amharic-to-Tigrigna MT using a hybrid approach ,

 Bi-directional Ge’ez-Amharic machine translation

English to Amharic Neural Machine


Translation
Statement of Problem Cont’d
 In recent studies, NMT provides promising results than other MT
approaches

 Currently, the progress in technology is shifting from traditional MT


approaches to deep learning-based NMT approach

English to Amharic Neural Machine


Translation
Objectives of the Study
General Objective
The general objective of this study is to design and
develop English -to- Amharic Neural Machine
Translation system.

English to Amharic Neural Machine


Translation
Objectives Cont’d
Specific Objectives
 Reviewing related systems and literature.

 Developing parallel bilingual corpus for Amharic and English languages.

 Identifying the linguistic behaviors of Amharic and English languages.

 Designing a general architecture for English -to- Amharic Neural


Machine Translation.

 Develop prototype.

 Testing and evaluating the performance of the system.

English to Amharic Neural Machine


Translation
Scope and Limitation of Study
 English -to- Amharic Neural Machine Translation is designed to
perform Neural Machine Translation of texts written in English
text to Amharic.

 Because of the unavailability of the standardized corpus (corpus


ready for MT research purpose) for the dataset, the dataset we
are going to use for training and testing is mostly from religious
documents

English to Amharic Neural Machine


Translation
Methodology
Research Design

An experimental research design will be used for


model building, analysis, and testing

Materials and Tools


 Software tools that the researcher will be used are python as a

programing language with Tensor-Flow environment and notepad.


English to Amharic Neural Machine
Translation
Methodology Cont’d
Dataset collection

 The translation system we are going to develop


tries to generate translations using the English -
Amharic corpus based on neural network methods

 The sources for both languages are a holy bible,


educational documents, and other relevant
materials
English to Amharic Neural Machine
Translation
English to Amharic Neural Machine
Translation

You might also like