Morphological Analysis for Phrase-Based Statistical Machine Translation

1
Morphological Analysis for Phrase-Based
Statistical Machine Translation
HYP update - part1

• Luong Minh Thang
• WING group meeting 15 Aug, 2008

2
Agenda

• Introduction - what does my project title mean?
• Language pair
• English-Finnish challenges
• Related works
• Project direction

3
Introduction I phrase-based SMT

• Statistical derive statistical information from
  large data
• Phrase-base capture local constraints

0 1 2 3 4 5 6 7
NULL Mary did not slap the green witch
Source
Target
Maria no daba una botefada a la bruja verde
1 2 3 4 5 6 7 8 9
4
Introduction II - Morphology

• Morpheme minimal meaning-bearing unit
• machines machine s
• translation translate ion
• goalkeeper goal keeper
• English is a low-inflected language - simple
  morphological structure
• ? High-inflected languages are much complicated!

5
Introduction III high-inflected languages

• Concatenate chain of morphemes to form a word
• Finnish oppositio kansa n edusta ja
• (opposition people of represent -ative)
• opposition of parliarment member
• Turkish uygarlas,tiramadiklarimizdanmis,sinizcasi
  na
• (uygarlas, tiramadiklarimizdanmis,
  sinizcasina)
• (behaving) as if you are among those whom we
  could not cause to become civilized

This is a word!!!
6
Introduction IV Why morphological-aware SMT?

• Tackle the data sparseness problem
• (Statistics from 1.021.180 sentence pairs)
• Capture the relations among words

Type count Token count
English 105.144 121.442.173
Finnish 516.102 130.128.883
7
Language pair I our choice?
Vietnamese

• We chose English - Finnish as our main
  translation task

8
Language pair II why Finnish?

• Honestly, I dont know Finnish
• But because
• Available corpora
• Finnish is an agglutinative morphologically-comple
  x language, suitable for our project scope
• Investigate in translation from low to high
  inflected languages -gt an area to explore, yet
  hard !!!

9
English-Finnish challenges I many-to-one word
relationship

• Finnish uses suffixes to express grammatical
  relations and also to derive new words

Case Suffix English prep. Sample word form Translation of the sample
nominatiivi   - talo house
genetiivi -n of talon of (a) house
essiivi -na as talona as a house
inessiivi -ssa in talossa in (a) house
elatiivi -sta from (inside) talosta from (a) house
komitatiivi -ne- together (with) taloineni with my house(s)
(about 14-15 cases for nouns)
Not merely concatenating

• Many-to-one English-Finnish word relationship ?
  need word-morpheme correspondence

10
English-Finnish challenges II word order

• Word order is free in Finnish
• Pete rakastaa Annaa Pete loves Annaa (normal)
• Annaa Pete rakastaa emphasizes Annaa
• Rakastaa Pete Annaa emphasizes rakastaa Pete
  does love Anna
• Pete Annaa rakastaa stress on Pete
• Rakastaa Annaa Pete. not sound like a normal
  sentence, quite understandable.

11
English-Finnish challenges III surface form
generation

• After translating from English words ? Finnish
  morphemes, need a surface generation step
• oppositio kansa n edusta ja ?
  oppositiokansanedustaja
• What if missing morphemes or changes in morpheme
  order?
• ? Need a more error-tolerate surface recovery
  algorithm

12
Related works I low-to-high inflected languages

• Many works from high to low inflected languages,
  but very few works on the opposite direction,
  considered hard in (Koehn, 2005)
• (Yang Kirchhoff, 2006) Finnish-English,
  backoff
• (Oflazer Durgar El-Kahlout, 2006, 2007)
  English-Turkish, word-morpheme translation, then
  simply concatenating morphemes
• All use language-dependent tools syntactic
  knowledge TreeTager, Snowball stemmer

13
Related works II surface form recovery

• (Toutanova et. al., 2007, 2008) English-Russian,
  English-Arabic translate stem-to-stem
• predict inflection from stems using many
  different features (lexical, morphological, and
  syntactic)
• (Avramidis Koehn, 2008) English-Greek
• Use syntax to get the missing morphology,
  depending on the syntactic position
• Noun cases agreement and verb person conjugation
• ? Rely mostly on manual annotation data

14
Project direction

• Use language-independent tool (Morfessor), and
  based on the unannotated data only
• (i.e. no feature data or syntactical
  information)
• Work on a general surface-form recovery
• We would like to have a unified view of the
  transalation process separating low-low,
  low-high, high-low, high-high

We are at here
15
Reference I

• Chirs Dyer, 2007 http//www.ling.umd.edu/redpony/
  edinburgh.pdf
• Jurafsky, D., Martin, J. H. (2007). Speech and
  language processing book
• The Finnish language http//www.cs.tut.fi/jkorpel
  a/Finnish.html
• Yang Kirchhoff, 2006 Phrase-based backoff
  models for machine translation of highly
  inflected languages
• Oflazer Durgar El-Kahlout, 2006 Initial
  Explorations in English to Turkish Statistical
  Machine Translation

16
Reference II

• Oflazer Durgar El-Kahlout, 2007 Exploring
  different representational units in
  English-to-Turkish statistical machine
  translation
• Toutanova et. al., 2007 Generating complex
  morphology for machine translation
• Toutanova et. al., 2008 Applying morphology
  generation models to machine translation
• Avramidis Koehn, 2008 Enriching
  morphologically poor languages for statistical
  machine translation

17
Q A?
18
To be continued

• Thank you !!!