Statistical Translation Language Model

1. Statistical Translation Language Model Maryam Karimzadehgan mkarimz2@illinois.edu University of Illinois at Urbana-Champaign

2. Outline • Motivation & Background • Language model (LM) for IR • Smoothing methods for IR • Statistical Machine Translation – Cross-Lingual • Motivation • IBM Model 1 • Statistical Translation Language Model – Monolingual • Synthetic Queries • Mutual Information-based approach • Regularization of self-translation probabilities • Smoothing in Statistical Translation Language Model

3. Language Model … text ? mining ? assocation ? clustering ? … food ? … ? Which model would most likely have generated this query? … food ? nutrition ? healthy ? diet ? … The Basic LM Approach([Ponte & Croft 98], [Hiemstra & Kraaij 98], [Miller et al. 99]) Document Query = “data mining algorithms” Text mining paper Food nutrition paper

4. Doc LM Query likelihood d1 p(q| d1) p(q| d2) d2 p(q| dN) dN Ranking Docs by Query Likelihood d1 q d2 dN

5. Retrieval as LM Estimation • Document ranking based on query likelihood Document language model • Retrieval problem  Estimation of p(wi|d) • Smoothing is an important issue, and distinguishes different approaches

6. How to Estimate p(w|d)? • Simplest solution: Maximum Likelihood Estimator • P(w|d) = relative frequency of word w in d • What if a word doesn’t appear in the text? P(w|d)=0 • In general, what probability should we give a word that has not been observed? • If we want to assign non-zero probabilities to such words, we’ll have to discount the probabilities of observed words • This is what “smoothing” is about …

7. P(w) Max. Likelihood Estimate Smoothed LM w Language Model Smoothing

8. Smoothing Methods for IR (Zhai & Lafferty 01) • Method 1(Linear interpolation, Jelinek-Mercer): • Method 2 (Dirichlet Prior/Bayesian): ML estimate parameter parameter

9. Outline • Motivation & Background • Language model (LM) for IR • Smoothing methods for IR • Statistical Machine Translation – Cross-Lingual • Motivation • IBM Model 1 • Statistical Translation Language Model – Monolingual • Synthetic Queries • Mutual Information-based approach • Regularization of self-translation probabilities • Smoothing in Statistical Translation Language Model

10. A Brief History • Machine translation was one of the first applications envisioned for computers • Warren Weaver (1949): “I have a text in front of me which is written in Russian but I am going to pretend that it is really written in English and that it has been coded in some strange symbols. All I need to do is strip off the code in order to retrieve the information contained in the text.” • First demonstrated by IBM in 1954 with a basic word-for-word translation system

11. Interest in Machine Translation • Commercial interest: • U.S. has invested in MT for intelligence purposes • MT is popular on the web—it is the most used of Google’s special features • EU spends more than $1 billion on translation costs each year. • (Semi-)automated translation could lead to huge savings

12. Interest in Machine Translation • Academic interest: • One of the most challenging problems in NLP research • Requires knowledge from many NLP sub-areas, e.g., lexical semantics, parsing, morphological analysis, statistical modeling,… • Being able to establish links between two languages allows for transferring resources from one language to another

13. Word-Level Alignments • Given a parallel sentence pair we can link (align) words or phrases that are translations of each other:

14. Machine Translation -- Concepts • We are trying to model P(e|f) • I give you a French sentence • You give me back English • How are we going to model this? • The maximum likelihood estimation of P(e | f) is: freq(e,f)/freq(f). • Way too specific to get any reasonable frequencies! Vast majority of unseen data will have zero counts!

15. Machine Translation – Alternative way • We could use Bayes rule • Why using Bayes rule and not directly estimating p(e|f) ? Given a French sentence f, we could do a search for an e that maximizes p(e|f). It is important that our model for p(e|f) concentrates its probability as much as possible on well-formed English sentences. But it is not important that our model for P(f|e) concentrate its probability on well-formed French sentences.

16. Statistical Machine Translation Language Model Translation Model Decoder • The noisy channel model • Assumptions: • An English word can be aligned with multiple French words while each French word is aligned with at most one English word • Independence of the individual word-to-word translations f: French e: English |e|=l |f|=m

17. Estimation of Probabilities -- IBM Model 1 • Simplest of the IBM models. (There are 5 models) • Does not consider word order (bag-of-words approach) • Does not model one-to-many alignments • Computationally inexpensive • Useful for parameter estimations that are passed on to more elaborate models

18. IBM Model 1 • Three important components involved • Language model • Give the probability p(e). • Translation model • Estimate the Translation Probability p(f|e). • Decoder

19. IBM Model 1- Translation Model • Joint probability of P(F=f, E=e, A=a) where A is an alignment between two sentences. • Assume each French word has exactly one connection.

20. IBM Model 1- Translation Model • Assume, |e|=l and |f|=m, then the alignment can be represented by a series a = a1a2…am • Each alignment is between 0 and l such that if the word in position j of the French sentence is connected to the word in position iof the English sentence, then aj=iand if it not connected to any English word, then aj=0. • The alignment is determined by specifying the values of aj for j from 1 to m, each of which can take any value from 0 to l.

21. IBM Model 1 – Translation Model all possible alignments (the English word that a French word fjis aligned with) translation probability EM algorithm is used to estimate the translation probabilities.

22. Outline • Motivation & Background • Language model (LM) for IR • Smoothing methods for IR • Statistical Machine Translation – Cross-Lingual • Motivation • IBM Model 1 • Statistical Translation Language Model – Monolingual • Synthetic Queries • Mutual Information-based approach • Regularization of self-translation probabilities • Smoothing in Statistical Translation Language Model

23. The Problem of Vocabulary Gap Query = auto wash P(“auto”) P(“wash”) auto wash … d1 How to support inexact matching? {“car” , “vehicle”} == “auto” “buy” ==== “wash” auto buy … auto d2 P(“auto”) P(“wash”) car wash vehicle d3

24. Translation Language Models for IR [Berger & Lafferty 99] Query = auto wash “translate” auto wash … “auto” “auto” Query = car wash d1 “car” auto buy auto P(“auto” |d3)= p(“car”|d3) x pt(“auto”| “car”) + p(“vehicle”|d3) x pt(“auto”| “vehicle”) d2 How to estimate? P(“car”|d3) P(“auto”) P(“wash”) car wash vehicle Pt(“auto”| “car”) “car” d3 “auto” “vehicle” P(“vehicle”|d3) Pt(“auto”| “vehicle”)

25. Estimation of Translation Model: pt(w|u) • When relevance judgments are available, (q,d) serves as data to train the translation model • Without relevance judgments, we can use synthetic data [Berger & Lafferty 99], <title, body>[Jin et al. 02] Basic translation model Regular doc LM Translation model

26. Estimation of Translation Model– Synthetic Queries([Berger & Lafferty 99]) • Select words that are representative of a document. • Calculate a Mutual information for each word in a document: I(w,d) = p(w,d)log • Synthetic queries are sampled based on normalized mutual information. • The resulting (d,q) of documents and synthetic queries are used to estimate the probabilities using EM algorithm (IBM Model 1).

27. Estimation of Translation Model– Synthetic Queries Algorithm([Berger & Lafferty 99]) Limitations: Can’t translate into words not seen in the training queries Computational complexity Training data

28. A simpler and more efficient method for estimating pt(w|u) with higher coveragewas proposed in:M. Karimzadehgan and C. Zhai. Estimation of Statistical Translation Models Based on Mutual Information for Ad Hoc Information Retrieval. ACM SIGIR, pages 323-330, 2010

29. Estimation of Translation Model Based on Mutual Information 1. Calculate Mutual information for each pair of two words in the collection (measuring co-occurrences) 2. Normalize mutual information score to obtain a translation probability: presence/absence of word w in a document

30. Computation Detail XwXu N D1 0 0 D2 1 1 D3 1 0 …. … DN 0 0 Xw=1 Xu=1 Exploit index to speed up computation

31. Sample Translation Probabilities (AP90) p(w| “everest”) Mutual Information Synthetic Query

32. Regularizing Self-Translation Probability • Self-translation probability can be under-estimated An exact match would be counted less than an exact match • Solution: Interpolation with “1.0 self-translation” w = u • = 1  basic query likelihood model • = 0  original MI estimate

33. Query Likelihood and Translation Language Model • Document ranking based on query likelihood Document language model • Translation Language Model Do you see any problem?

34. Further Smoothing of Translation Model for Computing Query Likelihood • Linear interpolation (Jelinek-Mercer): • Bayesian interpolation (Dirichlet prior): pml(w|d) pml(w|d)

35. Experiment Design • MI vs. Synthetic query estimation • Data Sets: Associated Press (AP90) and San Jose Mercury News (SJMN) + TREC topics 51-100 • Relatively small data sets in order to compare our results with Synthetic queries in [Berger& Lafferty 99]. • MI Translation model vs. Basic query likelihood • Larger Data Sets: TREC7, TREC8 (plus AP90, SJMN) • TREC topics 351-400 for TREC7 and 401-450 for TREC8 • Additional issues • Regularization of self-translation? • Influence of smoothing on translation models? • Translation model + pseudo feedback?

36. Mutual information outperforms synthetic queries in both MAP and P@10 AP90 + queries 51-100, Dirichlet Prior Smoothing MI MI Syn. Query Syn. Query

37. Upper Bound Comparison of Mutual Information and Synthetic Queries Dirichlet Prior Smoothing JM Smoothing

38. Mutual information translation model outperforms basic query likelihood Dir. Prior Smoothing JM Smoothing

39. Translation model appears to need less collection smoothing than basic QL Translation model Basic query likelihood

40. Translation model and pseudo feedback exploit word co-occurrences differently JM Smoothing Smoothed Translation Model Query model from pseudo FB

41. Regularization of self-translation is beneficial AP Data Set, DirichletPrior

42. Summary • Statistical Translation language model are effective for bridging the vocabulary gap. • Mutual information is more effective and more efficient than synthetic queries for estimating translation model probabilities. • Regularization of self-translation is beneficial • Translation model outperforms basic query likelihood on small and large collections and is more robust • Translation model and pseudo feedback exploit word co-occurrences differently and can be combined to further improve performance

43. References • [1] A. Berger and J. Lafferty. Information Retrieval as StatisticalTranslation. ACM SIGIR, pages 222–229, 1999. • [2] P. Brown, S. A. D. Pietra, V. J. D. Pietra, and R. Mercer. The mathematics of statistical machine translation: Parameter estimation. Computational Linguistics, 19(2):263–311, 1993. • [3] M. Karimzadehgan and C. Zhai. Estimation of Statistical Translation Models Based on Mutual Information for Ad Hoc Information Retrieval. ACM SIGIR, pages 323-330, 2010.