JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD

Course Objectives:

• 1. To teach students machine translation approaches.
• 2. To evaluate the performance of machine translation Systems.
• 3. To develop translation models for Indian Languages.

Course Outcomes

Upon the Successful Completion of the Course, the Students would be able to:

• 1. Understand machine translation approaches.
• 2. Apply and assess manual and automatic evaluation methods for machine translation.
• 3. Build machine translation model using existing tools for machine translation.

UNIT - I

Introduction to Machine Translation, MT Approaches: vauquois Triangle, Three major paradigms of Machine Translation, MT Evaluation

UNIT - II

Learning Bilingual word Mappings: A Combinatorial Argument, Deeper look at one- one alignment, Heuristic based Computation of the VE *VF Table, Iterative Computation of the VE *VF Table, EM: Study of progress in Parameter values

UNIT - III

Phrase based Machine Translation: Need for phrase alignment, An example to illustrate phrase alignment technique, Phrase table, Mathematics of Phrase based SMT, Decoding, Moses.

UNIT - IV

Rule based Machine Translation (RBMT): Two kinds of RBMT: Interlingua and Transfer, Universal networking Language (UNL), UNL expressions as binary predicates, Interlingua and Word Knowledge, Translation using Interlingua, Details of english to UNL Conversion: with illustration, Transfer based MT.

UNIT - V

Example based Machine Translation: Essential steps of EBMT, EBMTs working, EBMT and case based reasoning, Text similarity computation, EBMT and Translation Memory, EBMT and SMT.

TEXT BOOK:

• 1. Pushpak Bhattacharyya, Machine Translation, CRC Press

REFERENCES:

• 1. Statistical Machine Translation by Philipp Koehn, Cambridge University Press.
• 2. Deep Learning by Ian Goodfellow, Yoshua Bengio and Aaron Courville, MIT Press.
• 3. Linguistic Fundamentals for Natural Language Processing by Emily Bender, Morgan & Claypool.