Presentation is loading. Please wait.

Presentation is loading. Please wait.

CS 188: Artificial Intelligence Spring 2007 Lecture 25: Machine Translation 4/24/2007 Srini Narayanan – ICSI and UC Berkeley.

Similar presentations


Presentation on theme: "CS 188: Artificial Intelligence Spring 2007 Lecture 25: Machine Translation 4/24/2007 Srini Narayanan – ICSI and UC Berkeley."— Presentation transcript:

1 CS 188: Artificial Intelligence Spring 2007 Lecture 25: Machine Translation 4/24/2007 Srini Narayanan – ICSI and UC Berkeley

2 Announcements  Assignment 7 is up.  Grid-world and robot crawler.  Due 5/3.  Extra Office Hours first two weeks of May  This week as usual Thursday 11-1 PM  5/2 extra (Tuesday 11-1 PM)  5/3 usual 11-1 PM  Next assignment (not graded) will be a final exam review.

3 Reinforcement Learning  What you should know  MDPs  Basics, discounted reward  Policy Evaluation  Bellman’s equation  Value iteration  Policy iteration  Reinforcement Learning  Adaptive Dynamic Programming  TD learning (Model-free)  Q Learning

4 Where we are  Past:  Basic Techniques of AI  Search, Representation, Uncertainty and Inference, Learning  Next  Applications  MT, NLU (this week)  Neural Computation, Perception (next week).  Today: Machine Translation (MT)  (Semi) Automatically translating text/speech from one language to another.

5 Translation is hard In a Bucharest hotel lobby. The lift is being fixed for the next day. During that time we regret that you will be unbearable. In a Paris hotel elevator: Please leave your values at the front desk. In a hotel in Athens: Visitors are expected to complain at the office between the hours of 9 and 11 a.m. daily. In a Japanese hotel: You are invited to take advantage of the chambermaid. In the lobby of a Moscow hotel across from a Russian Orthodox monastery: You are welcome to visit the cemetery where famous Russian and Soviet composers, artists, and writers are buried daily except Thursday.

6 MT History  1946 (Pre-AI) Booth and Weaver discuss MT at Rockefeller foundation in New York;  idea of dictionary-based direct translation  1949 Weaver memorandum popularized idea  1952 all 18 MT researchers in world meet at MIT  1954 IBM/Georgetown Demo Russian-English MT  lots of labs take up MT

7 Early translation problems  English to Russian to English  The spirit is willing but the flesh is weak.  The vodka is good but the meat is rotten.

8 History of MT: Pessimism  1959/1960: Bar-Hillel “Report on the state of MT in US and GB”  Argued FAHQT too hard (semantic ambiguity, etc)  Should work on semi-automatic instead of automatic  His argument Little John was looking for his toy box. Finally, he found it. The box was in the pen. John was very happy.  Only human knowledge let’s us know that ‘playpens’ are bigger than boxes, but ‘writing pens’ are smaller  His claim: we would have to encode all of human knowledge

9 History of MT  Systran (Babelfish) been used for 30 years  1970’s:  European focus in MT; mainly ignored in US  1980’s  ideas of using AI techniques in MT (KBMT, CMU)  1990’s  Commercial MT systems  Statistical MT (SMT), Speech-to-speech translation  2000’s  SMT matures to be an exciting AI technology  Well funded, high-payoff, can make a real difference.

10 Levels of Transfer Interlingua Semantic Structure Semantic Structure Syntactic Structure Syntactic Structure Word Structure Word Structure Source Text Target Text Semantic Composition Semantic Decomposition Semantic Analysis Semantic Generation Syntactic Analysis Syntactic Generation Morphological Analysis Morphological Generation Semantic Transfer Syntactic Transfer Direct (Vauquois triangle)

11 General Approaches  Rule-based approaches  Expert system style rewrite systems  Interlingua methods (analyze and generate)  Lexicons come from humans or dictionaries  Can be very fast, and can accumulate a lot of knowledge over time (e.g. Systran)  Statistical approaches  Noisy channel systems  Lower-level transfer  Lexicons discovered using parallel corpora  Require little human declaration of knowledge

12 What makes a good translation  Translators often talk about two factors we want to maximize:  Faithfulness or fidelity  How close is the meaning of the translation to the meaning of the original  (Even better: does the translation cause the reader to draw the same inferences as the original would have)  Fluency or naturalness  How natural the translation is, just considering its fluency in the target language

13 The Coding View  “One naturally wonders if the problem of translation could conceivably be treated as a problem in cryptography. When I look at an article in Russian, I say: ‘This is really written in English, but it has been coded in some strange symbols. I will now proceed to decode.’ ”  Warren Weaver (1955:18, quoting a letter he wrote in 1947)

14 MT System Components source P(e) e f decoder observed argmax P(e|f) = argmax P(f|e)P(e) e e ef best channel P(f|e) Language ModelTranslation Model Finds an English translation which is both fluent and semantically faithful to the French source

15 The Classic Language Model Word N-Grams Generative approach: w1 = START repeat until END is generated: produce word w2 according to a big table P(w2 | w1) w1 := w2 P(I saw water on the table) = P(I | START) * P(saw | I) * P(water | saw) * P(on | water) * P(the | on) * P(table | the) * P(END | table) Probabilities can be learned from online English text. w1w1 w2w2 w n-1 END START

16 Parallel Corpora  Parallel corpora (or bitexts)  Collection of source- target translation pairs  Main resource for learning a translation model  Either naturally occurring (e.g. parliamentary proceedings, news translation services) or commissioned

17 Building a Translation Model  Steps in building a simple statistical translation model  Match up words in training sentence pairs (word alignment)  Learn a lexicon from these alignments  Learn larger phrases What is the anticipated cost of collecting fees under the new proposal ? En vertu de les nouvelles propositions, quel est le coût prévu de perception de les droits ?

18 Centauri/Arcturan [Knight, 1997] 1a. ok-voon ororok sprok. 1b. at-voon bichat dat. 7a. lalok farok ororok lalok sprok izok enemok. 7b. wat jjat bichat wat dat vat eneat. 2a. ok-drubel ok-voon anok plok sprok. 2b. at-drubel at-voon pippat rrat dat. 8a. lalok brok anok plok nok. 8b. iat lat pippat rrat nnat. 3a. erok sprok izok hihok ghirok. 3b. totat dat arrat vat hilat. 9a. wiwok nok izok kantok ok-yurp. 9b. totat nnat quat oloat at-yurp. 4a. ok-voon anok drok brok jok. 4b. at-voon krat pippat sat lat. 10a. lalok mok nok yorok ghirok clok. 10b. wat nnat gat mat bat hilat. 5a. wiwok farok izok stok. 5b. totat jjat quat cat. 11a. lalok nok crrrok hihok yorok zanzanok. 11b. wat nnat arrat mat zanzanat. 6a. lalok sprok izok jok stok. 6b. wat dat krat quat cat. 12a. lalok rarok nok izok hihok mok. 12b. wat nnat forat arrat vat gat. Your assignment, translate this to Arcturan: farok crrrok hihok yorok clok kantok ok-yurp

19 Centauri/Arcturan [Knight, 1997] 1a. ok-voon ororok sprok. 1b. at-voon bichat dat. 7a. lalok farok ororok lalok sprok izok enemok. 7b. wat jjat bichat wat dat vat eneat. 2a. ok-drubel ok-voon anok plok sprok. 2b. at-drubel at-voon pippat rrat dat. 8a. lalok brok anok plok nok. 8b. iat lat pippat rrat nnat. 3a. erok sprok izok hihok ghirok. 3b. totat dat arrat vat hilat. 9a. wiwok nok izok kantok ok-yurp. 9b. totat nnat quat oloat at-yurp. 4a. ok-voon anok drok brok jok. 4b. at-voon krat pippat sat lat. 10a. lalok mok nok yorok ghirok clok. 10b. wat nnat gat mat bat hilat. 5a. wiwok farok izok stok. 5b. totat jjat quat cat. 11a. lalok nok crrrok hihok yorok zanzanok. 11b. wat nnat arrat mat zanzanat. 6a. lalok sprok izok jok stok. 6b. wat dat krat quat cat. 12a. lalok rarok nok izok hihok mok. 12b. wat nnat forat arrat vat gat. Your assignment, translate this to Arcturan: farok crrrok hihok yorok clok kantok ok-yurp

20 Centauri/Arcturan [Knight, 1997] 1a. ok-voon ororok sprok. 1b. at-voon bichat dat. 7a. lalok farok ororok lalok sprok izok enemok. 7b. wat jjat bichat wat dat vat eneat. 2a. ok-drubel ok-voon anok plok sprok. 2b. at-drubel at-voon pippat rrat dat. 8a. lalok brok anok plok nok. 8b. iat lat pippat rrat nnat. 3a. erok sprok izok hihok ghirok. 3b. totat dat arrat vat hilat. 9a. wiwok nok izok kantok ok-yurp. 9b. totat nnat quat oloat at-yurp. 4a. ok-voon anok drok brok jok. 4b. at-voon krat pippat sat lat. 10a. lalok mok nok yorok ghirok clok. 10b. wat nnat gat mat bat hilat. 5a. wiwok farok izok stok. 5b. totat jjat quat cat. 11a. lalok nok crrrok hihok yorok zanzanok. 11b. wat nnat arrat mat zanzanat. 6a. lalok sprok izok jok stok. 6b. wat dat krat quat cat. 12a. lalok rarok nok izok hihok mok. 12b. wat nnat forat arrat vat gat.

21 Centauri/Arcturan [Knight, 1997] 1a. ok-voon ororok sprok. 1b. at-voon bichat dat. 7a. lalok farok ororok lalok sprok izok enemok. 7b. wat jjat bichat wat dat vat eneat. 2a. ok-drubel ok-voon anok plok sprok. 2b. at-drubel at-voon pippat rrat dat. 8a. lalok brok anok plok nok. 8b. iat lat pippat rrat nnat. 3a. erok sprok izok hihok ghirok. 3b. totat dat arrat vat hilat. 9a. wiwok nok izok kantok ok-yurp. 9b. totat nnat quat oloat at-yurp. 4a. ok-voon anok drok brok jok. 4b. at-voon krat pippat sat lat. 10a. lalok mok nok yorok ghirok clok. 10b. wat nnat gat mat bat hilat. 5a. wiwok farok izok stok. 5b. totat jjat quat cat. 11a. lalok nok crrrok hihok yorok zanzanok. 11b. wat nnat arrat mat zanzanat. 6a. lalok sprok izok jok stok. 6b. wat dat krat quat cat. 12a. lalok rarok nok izok hihok mok. 12b. wat nnat forat arrat vat gat. Your assignment, translate this to Arcturan: farok crrrok hihok yorok clok kantok ok-yurp ???

22 Centauri/Arcturan [Knight, 1997] 1a. ok-voon ororok sprok. 1b. at-voon bichat dat. 7a. lalok farok ororok lalok sprok izok enemok. 7b. wat jjat bichat wat dat vat eneat. 2a. ok-drubel ok-voon anok plok sprok. 2b. at-drubel at-voon pippat rrat dat. 8a. lalok brok anok plok nok. 8b. iat lat pippat rrat nnat. 3a. erok sprok izok hihok ghirok. 3b. totat dat arrat vat hilat. 9a. wiwok nok izok kantok ok-yurp. 9b. totat nnat quat oloat at-yurp. 4a. ok-voon anok drok brok jok. 4b. at-voon krat pippat sat lat. 10a. lalok mok nok yorok ghirok clok. 10b. wat nnat gat mat bat hilat. 5a. wiwok farok izok stok. 5b. totat jjat quat cat. 11a. lalok nok crrrok hihok yorok zanzanok. 11b. wat nnat arrat mat zanzanat. 6a. lalok sprok izok jok stok. 6b. wat dat krat quat cat. 12a. lalok rarok nok izok hihok mok. 12b. wat nnat forat arrat vat gat. Your assignment, translate this to Arcturan: farok crrrok hihok yorok clok kantok ok-yurp

23 Centauri/Arcturan [Knight, 1997] 1a. ok-voon ororok sprok. 1b. at-voon bichat dat. 7a. lalok farok ororok lalok sprok izok enemok. 7b. wat jjat bichat wat dat vat eneat. 2a. ok-drubel ok-voon anok plok sprok. 2b. at-drubel at-voon pippat rrat dat. 8a. lalok brok anok plok nok. 8b. iat lat pippat rrat nnat. 3a. erok sprok izok hihok ghirok. 3b. totat dat arrat vat hilat. 9a. wiwok nok izok kantok ok-yurp. 9b. totat nnat quat oloat at-yurp. 4a. ok-voon anok drok brok jok. 4b. at-voon krat pippat sat lat. 10a. lalok mok nok yorok ghirok clok. 10b. wat nnat gat mat bat hilat. 5a. wiwok farok izok stok. 5b. totat jjat quat cat. 11a. lalok nok crrrok hihok yorok zanzanok. 11b. wat nnat arrat mat zanzanat. 6a. lalok sprok izok jok stok. 6b. wat dat krat quat cat. 12a. lalok rarok nok izok hihok mok. 12b. wat nnat forat arrat vat gat. Your assignment, translate this to Arcturan: farok crrrok hihok yorok clok kantok ok-yurp

24 Centauri/Arcturan [Knight, 1997] 1a. ok-voon ororok sprok. 1b. at-voon bichat dat. 7a. lalok farok ororok lalok sprok izok enemok. 7b. wat jjat bichat wat dat vat eneat. 2a. ok-drubel ok-voon anok plok sprok. 2b. at-drubel at-voon pippat rrat dat. 8a. lalok brok anok plok nok. 8b. iat lat pippat rrat nnat. 3a. erok sprok izok hihok ghirok. 3b. totat dat arrat vat hilat. 9a. wiwok nok izok kantok ok-yurp. 9b. totat nnat quat oloat at-yurp. 4a. ok-voon anok drok brok jok. 4b. at-voon krat pippat sat lat. 10a. lalok mok nok yorok ghirok clok. 10b. wat nnat gat mat bat hilat. 5a. wiwok farok izok stok. 5b. totat jjat quat cat. 11a. lalok nok crrrok hihok yorok zanzanok. 11b. wat nnat arrat mat zanzanat. 6a. lalok sprok izok jok stok. 6b. wat dat krat quat cat. 12a. lalok rarok nok izok hihok mok. 12b. wat nnat forat arrat vat gat. Your assignment, translate this to Arcturan: farok crrrok hihok yorok clok kantok ok-yurp

25 Centauri/Arcturan [Knight, 1997] 1a. ok-voon ororok sprok. 1b. at-voon bichat dat. 7a. lalok farok ororok lalok sprok izok enemok. 7b. wat jjat bichat wat dat vat eneat. 2a. ok-drubel ok-voon anok plok sprok. 2b. at-drubel at-voon pippat rrat dat. 8a. lalok brok anok plok nok. 8b. iat lat pippat rrat nnat. 3a. erok sprok izok hihok ghirok. 3b. totat dat arrat vat hilat. 9a. wiwok nok izok kantok ok-yurp. 9b. totat nnat quat oloat at-yurp. 4a. ok-voon anok drok brok jok. 4b. at-voon krat pippat sat lat. 10a. lalok mok nok yorok ghirok clok. 10b. wat nnat gat mat bat hilat. 5a. wiwok farok izok stok. 5b. totat jjat quat cat. 11a. lalok nok crrrok hihok yorok zanzanok. 11b. wat nnat arrat mat zanzanat. 6a. lalok sprok izok jok stok. 6b. wat dat krat quat cat. 12a. lalok rarok nok izok hihok mok. 12b. wat nnat forat arrat vat gat. Your assignment, translate this to Arcturan: farok crrrok hihok yorok clok kantok ok-yurp ???

26 Centauri/Arcturan [Knight, 1997] 1a. ok-voon ororok sprok. 1b. at-voon bichat dat. 7a. lalok farok ororok lalok sprok izok enemok. 7b. wat jjat bichat wat dat vat eneat. 2a. ok-drubel ok-voon anok plok sprok. 2b. at-drubel at-voon pippat rrat dat. 8a. lalok brok anok plok nok. 8b. iat lat pippat rrat nnat. 3a. erok sprok izok hihok ghirok. 3b. totat dat arrat vat hilat. 9a. wiwok nok izok kantok ok-yurp. 9b. totat nnat quat oloat at-yurp. 4a. ok-voon anok drok brok jok. 4b. at-voon krat pippat sat lat. 10a. lalok mok nok yorok ghirok clok. 10b. wat nnat gat mat bat hilat. 5a. wiwok farok izok stok. 5b. totat jjat quat cat. 11a. lalok nok crrrok hihok yorok zanzanok. 11b. wat nnat arrat mat zanzanat. 6a. lalok sprok izok jok stok. 6b. wat dat krat quat cat. 12a. lalok rarok nok izok hihok mok. 12b. wat nnat forat arrat vat gat. Your assignment, translate this to Arcturan: farok crrrok hihok yorok clok kantok ok-yurp

27 Centauri/Arcturan [Knight, 1997] 1a. ok-voon ororok sprok. 1b. at-voon bichat dat. 7a. lalok farok ororok lalok sprok izok enemok. 7b. wat jjat bichat wat dat vat eneat. 2a. ok-drubel ok-voon anok plok sprok. 2b. at-drubel at-voon pippat rrat dat. 8a. lalok brok anok plok nok. 8b. iat lat pippat rrat nnat. 3a. erok sprok izok hihok ghirok. 3b. totat dat arrat vat hilat. 9a. wiwok nok izok kantok ok-yurp. 9b. totat nnat quat oloat at-yurp. 4a. ok-voon anok drok brok jok. 4b. at-voon krat pippat sat lat. 10a. lalok mok nok yorok ghirok clok. 10b. wat nnat gat mat bat hilat. 5a. wiwok farok izok stok. 5b. totat jjat quat cat. 11a. lalok nok crrrok hihok yorok zanzanok. 11b. wat nnat arrat mat zanzanat. 6a. lalok sprok izok jok stok. 6b. wat dat krat quat cat. 12a. lalok rarok nok izok hihok mok. 12b. wat nnat forat arrat vat gat. Your assignment, translate this to Arcturan: farok crrrok hihok yorok clok kantok ok-yurp process of elimination

28 Centauri/Arcturan [Knight, 1997] 1a. ok-voon ororok sprok. 1b. at-voon bichat dat. 7a. lalok farok ororok lalok sprok izok enemok. 7b. wat jjat bichat wat dat vat eneat. 2a. ok-drubel ok-voon anok plok sprok. 2b. at-drubel at-voon pippat rrat dat. 8a. lalok brok anok plok nok. 8b. iat lat pippat rrat nnat. 3a. erok sprok izok hihok ghirok. 3b. totat dat arrat vat hilat. 9a. wiwok nok izok kantok ok-yurp. 9b. totat nnat quat oloat at-yurp. 4a. ok-voon anok drok brok jok. 4b. at-voon krat pippat sat lat. 10a. lalok mok nok yorok ghirok clok. 10b. wat nnat gat mat bat hilat. 5a. wiwok farok izok stok. 5b. totat jjat quat cat. 11a. lalok nok crrrok hihok yorok zanzanok. 11b. wat nnat arrat mat zanzanat. 6a. lalok sprok izok jok stok. 6b. wat dat krat quat cat. 12a. lalok rarok nok izok hihok mok. 12b. wat nnat forat arrat vat gat. Your assignment, translate this to Arcturan: farok crrrok hihok yorok clok kantok ok-yurp cognate?

29 Your assignment, put these words in order: { jjat, arrat, mat, bat, oloat, at-yurp } Centauri/Arcturan [Knight, 1997] 1a. ok-voon ororok sprok. 1b. at-voon bichat dat. 7a. lalok farok ororok lalok sprok izok enemok. 7b. wat jjat bichat wat dat vat eneat. 2a. ok-drubel ok-voon anok plok sprok. 2b. at-drubel at-voon pippat rrat dat. 8a. lalok brok anok plok nok. 8b. iat lat pippat rrat nnat. 3a. erok sprok izok hihok ghirok. 3b. totat dat arrat vat hilat. 9a. wiwok nok izok kantok ok-yurp. 9b. totat nnat quat oloat at-yurp. 4a. ok-voon anok drok brok jok. 4b. at-voon krat pippat sat lat. 10a. lalok mok nok yorok ghirok clok. 10b. wat nnat gat mat bat hilat. 5a. wiwok farok izok stok. 5b. totat jjat quat cat. 11a. lalok nok crrrok hihok yorok zanzanok. 11b. wat nnat arrat mat zanzanat. 6a. lalok sprok izok jok stok. 6b. wat dat krat quat cat. 12a. lalok rarok nok izok hihok mok. 12b. wat nnat forat arrat vat gat. zero fertility

30 Clients do not sell pharmaceuticals in Europe => Clientes no venden medicinas en Europa It’s Really Spanish/English 1a. Garcia and associates. 1b. Garcia y asociados. 7a. the clients and the associates are enemies. 7b. los clients y los asociados son enemigos. 2a. Carlos Garcia has three associates. 2b. Carlos Garcia tiene tres asociados. 8a. the company has three groups. 8b. la empresa tiene tres grupos. 3a. his associates are not strong. 3b. sus asociados no son fuertes. 9a. its groups are in Europe. 9b. sus grupos estan en Europa. 4a. Garcia has a company also. 4b. Garcia tambien tiene una empresa. 10a. the modern groups sell strong pharmaceuticals. 10b. los grupos modernos venden medicinas fuertes. 5a. its clients are angry. 5b. sus clientes estan enfadados. 11a. the groups do not sell zenzanine. 11b. los grupos no venden zanzanina. 6a. the associates are also angry. 6b. los asociados tambien estan enfadados. 12a. the small groups are not modern. 12b. los grupos pequenos no son modernos.

31 Statistical Machine Translation … la maison … la maison bleue … la fleur … … the house … the blue house … the flower … All word alignments equally likely All P(french-word | english-word) equally likely

32 Statistical Machine Translation … la maison … la maison bleue … la fleur … … the house … the blue house … the flower … “la” and “the” observed to co-occur frequently, so P(la | the) is increased.

33 Statistical Machine Translation … la maison … la maison bleue … la fleur … … the house … the blue house … the flower … “house” co-occurs with both “la” and “maison”, but P(maison | house) can be raised without limit, to 1.0, while P(la | house) is limited because of “the” (pigeonhole principle)

34 Statistical Machine Translation … la maison … la maison bleue … la fleur … … the house … the blue house … the flower … settling down after another iteration

35 Statistical Machine Translation … la maison … la maison bleue … la fleur … … the house … the blue house … the flower … Inherent hidden structure revealed by EM training! For details, see: “ A Statistical MT Tutorial Workbook” (Knight, 1999). “The Mathematics of Statistical Machine Translation” (Brown et al, 1993) Software: GIZA++

36 Decoding  Now we have a phrase table:  A huge list of translation phrases (e.g. 1M phrases)  Each phrase has a probability P(f|e)  When we see a new input sentence:  Grow a translation left to right  Extend translation using known phrases  Also multiply by language model score

37 The Pharaoh Decoder  Probabilities at each step include LM and TM

38 Recent Progress in Statistical MT insistent Wednesday may recurred her trips to Libya tomorrow for flying Cairo 6-4 ( AFP ) - an official announced today in the Egyptian lines company for flying Tuesday is a company " insistent for flying " may resumed a consideration of a day Wednesday tomorrow her trips to Libya of Security Council decision trace international the imposed ban comment. And said the official " the institution sent a speech to Ministry of Foreign Affairs of lifting on Libya air, a situation her receiving replying are so a trip will pull to Libya a morning Wednesday ". Egyptair Has Tomorrow to Resume Its Flights to Libya Cairo 4-6 (AFP) - said an official at the Egyptian Aviation Company today that the company egyptair may resume as of tomorrow, Wednesday its flights to Libya after the International Security Council resolution to the suspension of the embargo imposed on Libya. " The official said that the company had sent a letter to the Ministry of Foreign Affairs, information on the lifting of the air embargo on Libya, where it had received a response, the first take off a trip to Libya on Wednesday morning " slide from C. Wayne, DARPA

39 Statistical Machine Translation … la maison … la maison bleue … la fleur … … the house … the blue house … the flower … P(juste | fair) = P(juste | correct) = P(juste | right) = … new French sentence Possible English translations, to be rescored by language model

40

41 What is MT not (yet) good for?  Really hard stuff  Literature  Natural spoken speech (meetings, court reporting)  Really important stuff  Medical translation in hospitals, 911

42 What is MT good for?  Tasks for which a rough translation is fine  Web pages,  Multilingual Speech-based queries  Tasks for which MT can be post-edited  MT as first pass  “Computer-aided human translation”  Tasks in sublanguage domains where high-quality MT is possible

43 The next five years  Bootstrapping Resources  Trying to design better learning methods to work from scarce data (see Knight 2003, Plauche et al 2007)  Germann and the ISI experiment in Tamil  MT in a month  100K tokens achieved tolerable performance in 2002  Including Syntactic/Semantic Information in SMT  Markup on the Web  Multi-lingual Lexical resources  WordNet PropBank FrameNet  Combining MT methods

44

45 Pos LanguageFamilyScript(s) Used Speakers Where Spoken (Major) 1MandarinSino-TibetanChinese Characters1051China, Malaysia, Taiwan 2EnglishIndo-EuropeanLatin510USA, UK, Australia, Canada, New Zealand 3HindiIndo-EuropeanDevanagari490North and Central India 4SpanishIndo-EuropeanLatin425The Americas, Spain 5ArabicAfro-AsiaticArabic255Middle East, Arabia, North Africa 6RussianIndo-EuropeanCyrillic254Russia, Central Asia 7PortugueseIndo-EuropeanLatin218Brazil, Portugal, Southern Africa 8BengaliIndo-EuropeanBengali215Bangladesh, Eastern India 9IndonesianMalayoPolynesianLatin175Indonesia, Malaysia, Singapore 10FrenchIndo-EuropeanLatin130France, Canada, West Africa, Central Africa 11JapaneseAltaicChinese Characters and 2 Japanese Alphabets127Japan 12GermanIndo-EuropeanLatin123Germany, Austria, Central Europe 13Farsi (Persian)Indo-EuropeanNastaliq110Iran, Afghanistan, Central Asia 14UrduIndo-EuropeanNastaliq104Pakistan, India 15PunjabiIndo-EuropeanGurumukhi103Pakistan, India 16VietnameseAustroasiaticBased on Latin86Vietnam, China 17TamilDravidianTamil78Southern India, Sri Lanka, Malyasia 18WuSino-TibetanChinese Characters77China 19JavaneseMalayo-PolynesianJavanese76Indonesia 20TurkishAltaicLatin75Turkey, Central Asia 21TeluguDravidianTelugu74Southern India 22KoreanAltaicHangul72Korean Peninsula 23MarathiIndo-EuropeanDevanagari71Western India 24ItalianIndo-EuropeanLatin61Italy, Central Europe 25ThaiSino-TibetanThai60Thailand, Laos 26CantoneseSino-TibetanChinese Characters55Southern China 27GujaratiIndo-EuropeanGujarati47Western India, Kenya 28PolishIndo-EuropeanLatin46Poland, Central Europe 29KannadaDravidianKannada44Southern India 30BurmeseSino-TibetanBurmese42Myanmar

46 Top Ten Internet Languages

47 MT in Developing Countries Traditional Rec Community Rec

48 Related Berkeley work at TIER  Kiosks / Livelihood  Cellphones for pricing in rural Rwandan coffee markets  Computers and livelihood development in urban slums in Brazil  E-literacy / Entrepreneurship in rural Kerala  Education  Studies of social impacts of Computer Aided Learning in rural areas  Observations of shared computer usage among children in resource strapped areas  Telemedicine  Long-distance diagnosis using b  Teaching  ‘Technology and Development’ graduate class design (see reader/syllabus)  Conference  First peer-reviewed IEEE/ACM conference in series

49 URL bibliography  website.  website.  website.     browser.       WordNet Association.     HowNet.  multilingual semantic network.  project.  project.  project for  Japanese.  project for Spanish.  project.   of VerbNet and  FrameNet.  NomBank

50 References


Download ppt "CS 188: Artificial Intelligence Spring 2007 Lecture 25: Machine Translation 4/24/2007 Srini Narayanan – ICSI and UC Berkeley."

Similar presentations


Ads by Google