acl acl2011 acl2011-193 knowledge-graph by maker-knowledge-mining
Source: pdf
Author: Klaus Macherey ; Andrew Dai ; David Talbot ; Ashok Popat ; Franz Och
Abstract: Translating compounds is an important problem in machine translation. Since many compounds have not been observed during training, they pose a challenge for translation systems. Previous decompounding methods have often been restricted to a small set of languages as they cannot deal with more complex compound forming processes. We present a novel and unsupervised method to learn the compound parts and morphological operations needed to split compounds into their compound parts. The method uses a bilingual corpus to learn the morphological operations required to split a compound into its parts. Furthermore, monolingual corpora are used to learn and filter the set of compound part candidates. We evaluate our method within a machine translation task and show significant improvements for various languages to show the versatility of the approach.
Reference: text
sentIndex sentText sentNum sentScore
1 Since many compounds have not been observed during training, they pose a challenge for translation systems. [sent-7, score-0.319]
2 Previous decompounding methods have often been restricted to a small set of languages as they cannot deal with more complex compound forming processes. [sent-8, score-0.983]
3 We present a novel and unsupervised method to learn the compound parts and morphological operations needed to split compounds into their compound parts. [sent-9, score-2.342]
4 The method uses a bilingual corpus to learn the morphological operations required to split a compound into its parts. [sent-10, score-1.209]
5 Furthermore, monolingual corpora are used to learn and filter the set of compound part candidates. [sent-11, score-0.9]
6 1 Introduction A compound is a lexeme that consists of more than one stem. [sent-13, score-0.778]
7 Informally, a compound is a combination of two or more words that function as a single unit of meaning. [sent-14, score-0.797]
8 Some compounds are written as space-separated words, which are called open compounds (e. [sent-15, score-0.514]
9 hard drive), while others are written as single words, which are called closed compounds (e. [sent-17, score-0.288]
10 In this paper, we shall focus only on closed compounds because open compounds do not require further splitting. [sent-20, score-0.571]
11 The objective of compound splitting is to split a compound into its corresponding sequence of constituents. [sent-21, score-1.842]
12 If we look at how compounds are created from lexemes in the first place, we find that for some languages, compounds are formed by concatenating 1395 2University of Edinburgh 10 Crichton Street . [sent-22, score-0.598]
13 1 Case Studies Below, we look at splitting compounds from 3 different languages. [sent-29, score-0.431]
14 1 English Compound Splitting The word flowerpot can appear as a closed or open compound in English texts. [sent-34, score-0.913]
15 To automatically split the closed form we have to try out every split point and choose the split with minimal costs according to a cost function. [sent-35, score-0.525]
16 Since flowerpot consists of 9 characters, we have 8 possibilities to position split point n1 within the characters c1, . [sent-41, score-0.294]
17 2 German Compound Splitting The previous example covered a case where the compound is constructed by directly concatenating the compound parts. [sent-57, score-1.572]
18 To demonstrate, we look at the Ger- man compound Verkehrszeichen (traffic sign) which consists of the two nouns Verkehr (traffic) and Zeichen (sign). [sent-59, score-0.801]
19 Verkehrszeichen → Verkehrszeich + e + n Using the same procedure as described before, we can lookup the compound parts in a dictionary or determine their frequency from large text collections. [sent-68, score-1.025]
20 The interesting part here is the additional s morpheme, which is called a linking morpheme, because it combines the two compound parts to form the compound Verkehrszeichen. [sent-70, score-1.774]
21 If we have a list of all possible linking morphemes, we can hypothesize them between two ordinary compound parts. [sent-71, score-0.851]
22 3 Greek Compound Splitting The previous example required the insertion of a linking morpheme between two compound parts. [sent-74, score-0.852]
23 The Greek compound χαρτόκουτο (cardboard box) consists of the two parts χαρτί (paper) and κουτί (box). [sent-76, score-0.927]
24 To lookup the correct words, we must substitute the suffix of the compound part candidates with some other morphemes. [sent-78, score-0.875]
25 If we allow 1396 the compound part candidates to be transformed by some morphological operation, we can lookup the transformed compound parts in a dictionary or determine their frequencies in some large collection of Greek texts. [sent-79, score-2.004]
26 Then this yields the following compound part candidates: χαρτόκουτο χαρτόκουτο χαρτόκουτο → χ → χ → χ . [sent-81, score-0.806]
27 → → → χαρτί + κουτί χαρτίκουτ χαρτίκουτ + ο + ί ό/ί, g2: ο / ί : ό / ί g1 : g1 g2 : ο / ί Here, gk : s/t denotes the kth compound part which is obtained by replacing string s with string t in the original string, resulting in the transformed part gk. [sent-85, score-0.923]
28 2 Problems and Objectives Our goal is to design a language-independent compound splitter that is useful for machine translation. [sent-87, score-0.974]
29 The previous examples addressed the importance of a cost function that favors valid compound parts versus invalid ones. [sent-88, score-1.054]
30 Because it seems infeasible to list the morphological operations explic- itly, we want to find and extract those operations automatically in an unsupervised way and provide them as an additional knowledge source to the decompounding algorithm. [sent-92, score-0.554]
31 Another problem is how to evaluate the quality of the compound splitter. [sent-93, score-0.778]
32 One way is to compile for every language a large collection of compounds together with their valid splits and to measure the proportion of correctly split compounds. [sent-94, score-0.466]
33 While the training algorithm for our compound splitter shall be unsupervised, the evaluation data needs to be verified by human experts. [sent-96, score-1.0]
34 Since we are interested in improving machine translation and to circumvent the problem of explicitly annotating compounds, we evaluate the compound splitter within a machine translation task. [sent-97, score-1.098]
35 For example, the German wordDeutschland (Germany) could be split into two parts Deutsch (German) + Land (country). [sent-102, score-0.284]
36 To avoid overly eager splitting, we will compile a list ofnon-compounds in an unsupervised way that serves as an exception list for the compound splitter. [sent-105, score-0.973]
37 To summarize, we aim to solve the following problems: • • • • 2 Define a cost function that favors valid compound parts and rejects invalid ones. [sent-106, score-1.054]
38 Learn morphological operations, which is important for languages that have complex compound forming processes. [sent-107, score-0.976]
39 Apply compound splitting to machine translation to aid in translation ofcompounds that have not been seen in the bilingual training data. [sent-108, score-1.104]
40 Brown (2002) describes a corpus-driven approach for splitting compounds in a German-English translation task derived from a medical domain. [sent-111, score-0.47]
41 While the English text keeps the cognates as separate tokens, they are combined into compounds in the German text. [sent-113, score-0.314]
42 To split these compounds, the author compares both the German and the English cognates on a character level to find reasonable split points. [sent-114, score-0.352]
43 Koehn and Knight (2003) present a frequencybased approach to compound splitting for German. [sent-117, score-0.929]
44 The compound parts and their frequencies are es- timated from a monolingual corpus. [sent-118, score-0.996]
45 The authors allow only two linking morphemes between compound parts and a few letters that can be dropped. [sent-120, score-1.024]
46 Garera and Yarowsky (2008) propose an approach to translate compounds without the need for bilingual training texts. [sent-122, score-0.327]
47 The compound splitting procedure mainly follows the approach from (Brown, 2002) and (Koehn and Knight, 2003), so the emphasis is put on finding correct translations for compounds. [sent-123, score-0.958]
48 To accomplish this, the authors use crosslanguage compound evidence obtained from bilingual dictionaries. [sent-124, score-0.829]
49 (2008b) describe a state-of-theart German compound splitter that is particularly robust with respect to noise and spelling errors. [sent-128, score-0.974]
50 The compound splitter is trained on monolingual data. [sent-129, score-1.019]
51 Besides applying frequency and probability-based methods, the authors also take the mutual information of compound parts into account. [sent-130, score-0.958]
52 A very frequent glue morpheme like -es- is not listed, while glue morphemes like -k- and -h- rank very high, although they are invalid glue morphemes for German. [sent-132, score-0.32]
53 authors look for compound parts that occur in different anchor texts pointing to the same document. [sent-134, score-0.979]
54 (2008a) apply this compound splitter on various other Germanic languages. [sent-137, score-0.974]
55 Dyer (2009) applies a maximum entropy model of compound splitting to generate segmentation lattices that serve as input to a translation system. [sent-138, score-1.007]
56 3 Compound Splitting Algorithm In this section, we describe the underlying optimization problem and the algorithm used to split a token into its compound parts. [sent-141, score-0.942]
57 Starting from Bayes' decision rule, we develop the Bellman equation and formulate a dynamic programming-based algorithm that takes a word as input and outputs the constituent compound parts. [sent-142, score-0.817]
58 We discuss the procedure used to extract compound parts from monolingual texts and to learn the morphological operations using bilingual corpora. [sent-143, score-1.326]
59 If we penalize each split with a constant split penalty ξ, and make the probability independent of the number of splits K, we arrive at the following decision rule: w = c1N → (Kˆ,ˆ n1Kˆ, ˆ g1Kˆ) ∏K = Kar,gn0Km,ga1Kx{ξK·k∏=1p(cnnkk−1+1,nk−1,gk)} (6) 3. [sent-158, score-0.332]
60 This yields the following recursive equation: Q(c1j) = mnk,agxk{ξ · Q(c1nk−1)· · p(cnnk+1, gk)} nk−1, (7) Algorithm 1Compound splitting Input: input word w = c1N Output: compound parts Q(0) = 0 QQ((01)) == ·0 · · = Q(N) = ∞ fQor(1 1i) = = 0 ·,· . [sent-163, score-1.078]
61 , en) pairs, which are potential compound candidates together with their English translations. [sent-176, score-0.809]
62 To mitigate this, thresholds such as minimum edit dis- tance between the potential compound and its parts, minimum co-occurrence frequencies for the selected bilingual phrase pairs and minimum source and target word lengths are used to reduce the noise at the expense of finding fewer compounds. [sent-180, score-0.932]
63 We then calculate the distance between all permutations of the parts and the original compound and choose the one with the lowest distance and highest transla- tion probability: Überweisung Betrag. [sent-197, score-0.927]
64 2 Monolingual Extraction of Compound Parts The most important knowledge source required for Algorithm 1 is a word-frequency list of compound parts that is used to compute the split costs. [sent-199, score-1.119]
65 1 is useful for learning morphological operations, but it is not sufficient to extract an exhaustive list of compound parts. [sent-202, score-0.934]
66 The output is a table that contains preliminary compound parts together with their respective counts for each language. [sent-211, score-0.944]
67 2 Phase 2: Filtering pass In the second pass, the compound part vocabulary is further reduced and filtered. [sent-214, score-0.921]
68 We generate a LM vocabulary based on arbitrary web texts for each language and build a compound splitter based on the vocabulary list that was generated in phase 1. [sent-215, score-1.15]
69 We now try to split every word of the web LM vocabulary based on the compound splitter model from phase 1. [sent-216, score-1.186]
70 For the compound parts that occur in the compound splitter output, we determine how often each compound part was used and output only those compound parts whose frequency exceed a predefined threshold n. [sent-217, score-3.665]
71 3 Example Suppose we have the following word frequencies output from pass 1: floor 10k poll 4k flow 9k pot 5k flower 15k potter 20k In pass 2, we observe the word flowerpot. [sent-219, score-0.291]
72 With the above list, the only compound parts used are flower and pot. [sent-220, score-1.001]
73 This filtering pass has the advantage of outputting only those compound part candidates 1400 which were actually used to split words from web texts. [sent-222, score-1.066]
74 Another advantage is that since the set of parts output in the first pass may contain a high number of compounds, the filter is able to remove a large number of these compounds by examining relative frequencies. [sent-224, score-0.532]
75 In our experiments, we have assumed that compound part frequencies are higher than the compound frequency and so remove words from the part list that can themselves be split and have a relatively high frequency. [sent-225, score-1.834]
76 Finally, after removing the low frequency compound parts, we obtain the final compound splitter vocabulary. [sent-226, score-1.783]
77 • • • The split penalty ξ penalizes each compound part to avoid eager splitting. [sent-234, score-1.033]
78 The cost for each compound part gk is computed as −log C(gk), where C(gk) is the unipgurtaemd caosu −ntl fogorC gk obtained from the news LM word frequency list. [sent-235, score-1.034]
79 The publicly available Europarl corpus is not suitable for demonstrating the utility of compound splitting because there are few unseen compounds in the test section of the Europarl corpus. [sent-242, score-1.186]
80 1is not purely restricted to the case of decompounding, we found that many morphological operations emitted by this procedure reflect morphological variations that are not directly linked to compounding, but caused by inflections. [sent-253, score-0.398]
81 To generate the language-dependent lists of compound parts, we used language model vocabulary lists5 generated from news texts for different languages as seeds for the first pass. [sent-254, score-0.997]
82 For the second pass, we used the lists generated in the first pass together with the learned morphological operations to construct a preliminary compound splitter. [sent-257, score-1.186]
83 We then generated vocabulary lists for monolingual web texts and applied the preliminary compound splitter onto this list. [sent-258, score-1.189]
84 operations Table 2: Examples of morphological operations that were extracted from bilingual corpora. [sent-266, score-0.417]
85 compound parts were collected and sorted according to their frequencies. [sent-267, score-0.943]
86 Those which were used at least 2 times were kept in the final compound parts lists. [sent-268, score-0.952]
87 Table 3 reports the number of compound parts kept after each pass. [sent-269, score-0.952]
88 Using the preliminary compound splitter in the second filter step resulted in 73K compound part entries. [sent-273, score-1.846]
89 The finally obtained compound splitter was integrated into the preprocessing pipeline of a stateof-the-art statistical phrase-based machine translation system that works similar to the Moses decoder (Koehn et al. [sent-274, score-1.036]
90 By applying the compound splitter during both training and decoding we ensured that source language tokens were split in the same way. [sent-276, score-1.157]
91 While the compounding process for Ger- manic languages is rather simple and requires only a 1402 few linking morphemes, compounds used in Uralic languages have a richer morphology. [sent-281, score-0.455]
92 We analyzed the number of remaining source characters in the baseline system and the system using compound splitting by counting the number of Greek characters in the translation output. [sent-284, score-1.126]
93 Because we do not know how many compounds are actually contained in the Greek source sentences6 and because the frequency of using compounds might vary across languages, we cannot expect the same performance gains across languages belonging to different language families. [sent-287, score-0.646]
94 2 BLEU points across translation systems and increases the compound parts vocabulary lists by up to 20%, which means that most of the gains can be achieved with simple concatenation. [sent-292, score-1.139]
95 765 02741) Table 3: Number of remaining compound parts for various languages after the first and second filter step. [sent-300, score-1.034]
96 SystemBLEU[%] w/o splittingBLEU[%] w splitting∆CI 95% Table 4: BLEU score results for various languages translated into English with and without compound splitting. [sent-301, score-0.836]
97 Table 5: Examples of translations into English with and without compound splitting. [sent-310, score-0.778]
98 6 Conclusion and Outlook We have presented a language-independent method for decompounding that improves translations for compounds that otherwise rarely occur in the bilingual training data. [sent-315, score-0.439]
99 We learned a set of morphological operations from a translation phrase table and determined suitable compound part candidates from monolingual data in a two pass process. [sent-316, score-1.266]
100 All knowledge sources were combined in a cost function that was applied in a compound splitter based on dynamic programming. [sent-319, score-1.064]
wordName wordTfidf (topN-words)
[('compound', 0.778), ('compounds', 0.257), ('splitter', 0.196), ('splitting', 0.151), ('parts', 0.149), ('split', 0.135), ('decompounding', 0.131), ('morphological', 0.124), ('operations', 0.121), ('flowerpot', 0.104), ('greek', 0.086), ('pass', 0.077), ('flower', 0.074), ('verkehrszeichen', 0.074), ('nk', 0.072), ('lists', 0.069), ('translation', 0.062), ('gk', 0.062), ('german', 0.06), ('languages', 0.058), ('cognates', 0.057), ('morphemes', 0.056), ('characters', 0.055), ('eager', 0.052), ('bilingual', 0.051), ('filter', 0.049), ('cost', 0.048), ('glue', 0.048), ('lexemes', 0.045), ('alfonseca', 0.045), ('monolingual', 0.045), ('cnnkk', 0.044), ('uralic', 0.044), ('germanic', 0.043), ('costs', 0.041), ('linking', 0.041), ('compounding', 0.041), ('pot', 0.039), ('lookup', 0.038), ('vocabulary', 0.038), ('pr', 0.034), ('entries', 0.034), ('morpheme', 0.033), ('bleu', 0.033), ('list', 0.032), ('lm', 0.032), ('frequency', 0.031), ('closed', 0.031), ('candidates', 0.031), ('invalid', 0.031), ('berweisung', 0.03), ('verkehr', 0.03), ('zeichen', 0.03), ('valid', 0.029), ('procedure', 0.029), ('texts', 0.029), ('token', 0.029), ('part', 0.028), ('transformed', 0.027), ('splits', 0.027), ('bilac', 0.026), ('slaven', 0.026), ('shall', 0.026), ('source', 0.025), ('points', 0.025), ('kept', 0.025), ('character', 0.025), ('news', 0.025), ('garera', 0.024), ('punctuations', 0.024), ('frequencies', 0.024), ('tokens', 0.023), ('exception', 0.023), ('dynamic', 0.023), ('look', 0.023), ('phase', 0.022), ('europarl', 0.022), ('traffic', 0.021), ('avoid', 0.021), ('enrique', 0.021), ('koehn', 0.02), ('penalty', 0.019), ('translate', 0.019), ('named', 0.019), ('function', 0.019), ('compile', 0.018), ('minimum', 0.018), ('gains', 0.018), ('digits', 0.017), ('preliminary', 0.017), ('overly', 0.017), ('discarding', 0.017), ('entities', 0.017), ('web', 0.017), ('segmentations', 0.017), ('decision', 0.016), ('concatenating', 0.016), ('sorted', 0.016), ('forming', 0.016), ('lattices', 0.016)]
simIndex simValue paperId paperTitle
same-paper 1 1.0000005 193 acl-2011-Language-independent compound splitting with morphological operations
Author: Klaus Macherey ; Andrew Dai ; David Talbot ; Ashok Popat ; Franz Och
Abstract: Translating compounds is an important problem in machine translation. Since many compounds have not been observed during training, they pose a challenge for translation systems. Previous decompounding methods have often been restricted to a small set of languages as they cannot deal with more complex compound forming processes. We present a novel and unsupervised method to learn the compound parts and morphological operations needed to split compounds into their compound parts. The method uses a bilingual corpus to learn the morphological operations required to split a compound into its parts. Furthermore, monolingual corpora are used to learn and filter the set of compound part candidates. We evaluate our method within a machine translation task and show significant improvements for various languages to show the versatility of the approach.
2 0.24817841 247 acl-2011-Pre- and Postprocessing for Statistical Machine Translation into Germanic Languages
Author: Sara Stymne
Abstract: In this thesis proposal Ipresent my thesis work, about pre- and postprocessing for statistical machine translation, mainly into Germanic languages. I focus my work on four areas: compounding, definite noun phrases, reordering, and error correction. Initial results are positive within all four areas, and there are promising possibilities for extending these approaches. In addition Ialso focus on methods for performing thorough error analysis of machine translation output, which can both motivate and evaluate the studies performed.
3 0.20037305 249 acl-2011-Predicting Relative Prominence in Noun-Noun Compounds
Author: Taniya Mishra ; Srinivas Bangalore
Abstract: There are several theories regarding what influences prominence assignment in English noun-noun compounds. We have developed corpus-driven models for automatically predicting prominence assignment in noun-noun compounds using feature sets based on two such theories: the informativeness theory and the semantic composition theory. The evaluation of the prediction models indicate that though both of these theories are relevant, they account for different types of variability in prominence assignment.
4 0.10009994 310 acl-2011-Translating from Morphologically Complex Languages: A Paraphrase-Based Approach
Author: Preslav Nakov ; Hwee Tou Ng
Abstract: We propose a novel approach to translating from a morphologically complex language. Unlike previous research, which has targeted word inflections and concatenations, we focus on the pairwise relationship between morphologically related words, which we treat as potential paraphrases and handle using paraphrasing techniques at the word, phrase, and sentence level. An important advantage of this framework is that it can cope with derivational morphology, which has so far remained largely beyond the capabilities of statistical machine translation systems. Our experiments translating from Malay, whose morphology is mostly derivational, into English show signif- icant improvements over rivaling approaches based on five automatic evaluation measures (for 320,000 sentence pairs; 9.5 million English word tokens).
5 0.097305849 10 acl-2011-A Discriminative Model for Joint Morphological Disambiguation and Dependency Parsing
Author: John Lee ; Jason Naradowsky ; David A. Smith
Abstract: Most previous studies of morphological disambiguation and dependency parsing have been pursued independently. Morphological taggers operate on n-grams and do not take into account syntactic relations; parsers use the “pipeline” approach, assuming that morphological information has been separately obtained. However, in morphologically-rich languages, there is often considerable interaction between morphology and syntax, such that neither can be disambiguated without the other. In this paper, we propose a discriminative model that jointly infers morphological properties and syntactic structures. In evaluations on various highly-inflected languages, this joint model outperforms both a baseline tagger in morphological disambiguation, and a pipeline parser in head selection.
6 0.086897105 75 acl-2011-Combining Morpheme-based Machine Translation with Post-processing Morpheme Prediction
7 0.08638303 318 acl-2011-Unsupervised Bilingual Morpheme Segmentation and Alignment with Context-rich Hidden Semi-Markov Models
8 0.08610896 163 acl-2011-Improved Modeling of Out-Of-Vocabulary Words Using Morphological Classes
9 0.083323926 241 acl-2011-Parsing the Internal Structure of Words: A New Paradigm for Chinese Word Segmentation
10 0.081321694 127 acl-2011-Exploiting Web-Derived Selectional Preference to Improve Statistical Dependency Parsing
11 0.080813944 16 acl-2011-A Joint Sequence Translation Model with Integrated Reordering
12 0.069810919 259 acl-2011-Rare Word Translation Extraction from Aligned Comparable Documents
13 0.068036914 104 acl-2011-Domain Adaptation for Machine Translation by Mining Unseen Words
14 0.066668361 6 acl-2011-A Comprehensive Dictionary of Multiword Expressions
15 0.057831571 202 acl-2011-Learning Hierarchical Translation Structure with Linguistic Annotations
16 0.055943668 157 acl-2011-I Thou Thee, Thou Traitor: Predicting Formal vs. Informal Address in English Literature
17 0.055690512 90 acl-2011-Crowdsourcing Translation: Professional Quality from Non-Professionals
18 0.054611105 331 acl-2011-Using Large Monolingual and Bilingual Corpora to Improve Coordination Disambiguation
19 0.053825237 333 acl-2011-Web-Scale Features for Full-Scale Parsing
20 0.052858781 44 acl-2011-An exponential translation model for target language morphology
topicId topicWeight
[(0, 0.146), (1, -0.073), (2, 0.03), (3, 0.034), (4, -0.003), (5, -0.012), (6, 0.078), (7, -0.035), (8, 0.041), (9, 0.055), (10, -0.019), (11, -0.006), (12, -0.055), (13, -0.059), (14, 0.075), (15, -0.046), (16, -0.013), (17, -0.01), (18, 0.053), (19, 0.019), (20, 0.026), (21, 0.012), (22, 0.011), (23, -0.145), (24, -0.007), (25, 0.007), (26, -0.006), (27, 0.022), (28, -0.056), (29, 0.018), (30, 0.061), (31, 0.007), (32, 0.116), (33, -0.09), (34, 0.033), (35, 0.104), (36, 0.001), (37, 0.02), (38, -0.157), (39, 0.163), (40, -0.088), (41, -0.126), (42, -0.002), (43, -0.084), (44, 0.065), (45, 0.056), (46, 0.104), (47, -0.112), (48, -0.035), (49, -0.012)]
simIndex simValue paperId paperTitle
same-paper 1 0.94076335 193 acl-2011-Language-independent compound splitting with morphological operations
Author: Klaus Macherey ; Andrew Dai ; David Talbot ; Ashok Popat ; Franz Och
Abstract: Translating compounds is an important problem in machine translation. Since many compounds have not been observed during training, they pose a challenge for translation systems. Previous decompounding methods have often been restricted to a small set of languages as they cannot deal with more complex compound forming processes. We present a novel and unsupervised method to learn the compound parts and morphological operations needed to split compounds into their compound parts. The method uses a bilingual corpus to learn the morphological operations required to split a compound into its parts. Furthermore, monolingual corpora are used to learn and filter the set of compound part candidates. We evaluate our method within a machine translation task and show significant improvements for various languages to show the versatility of the approach.
2 0.68004119 249 acl-2011-Predicting Relative Prominence in Noun-Noun Compounds
Author: Taniya Mishra ; Srinivas Bangalore
Abstract: There are several theories regarding what influences prominence assignment in English noun-noun compounds. We have developed corpus-driven models for automatically predicting prominence assignment in noun-noun compounds using feature sets based on two such theories: the informativeness theory and the semantic composition theory. The evaluation of the prediction models indicate that though both of these theories are relevant, they account for different types of variability in prominence assignment.
3 0.62555844 247 acl-2011-Pre- and Postprocessing for Statistical Machine Translation into Germanic Languages
Author: Sara Stymne
Abstract: In this thesis proposal Ipresent my thesis work, about pre- and postprocessing for statistical machine translation, mainly into Germanic languages. I focus my work on four areas: compounding, definite noun phrases, reordering, and error correction. Initial results are positive within all four areas, and there are promising possibilities for extending these approaches. In addition Ialso focus on methods for performing thorough error analysis of machine translation output, which can both motivate and evaluate the studies performed.
4 0.61203903 75 acl-2011-Combining Morpheme-based Machine Translation with Post-processing Morpheme Prediction
Author: Ann Clifton ; Anoop Sarkar
Abstract: This paper extends the training and tuning regime for phrase-based statistical machine translation to obtain fluent translations into morphologically complex languages (we build an English to Finnish translation system) . Our methods use unsupervised morphology induction. Unlike previous work we focus on morphologically productive phrase pairs – our decoder can combine morphemes across phrase boundaries. Morphemes in the target language may not have a corresponding morpheme or word in the source language. Therefore, we propose a novel combination of post-processing morphology prediction with morpheme-based translation. We show, using both automatic evaluation scores and linguistically motivated analyses of the output, that our methods outperform previously proposed ones and pro- vide the best known results on the EnglishFinnish Europarl translation task. Our methods are mostly language independent, so they should improve translation into other target languages with complex morphology. 1 Translation and Morphology Languages with rich morphological systems present significant hurdles for statistical machine translation (SMT) , most notably data sparsity, source-target asymmetry, and problems with automatic evaluation. In this work, we propose to address the problem of morphological complexity in an Englishto-Finnish MT task within a phrase-based translation framework. We focus on unsupervised segmentation methods to derive the morphological information supplied to the MT model in order to provide coverage on very large datasets and for languages with few hand-annotated 32 resources. In fact, in our experiments, unsupervised morphology always outperforms the use of a hand-built morphological analyzer. Rather than focusing on a few linguistically motivated aspects of Finnish morphological behaviour, we develop techniques for handling morphological complexity in general. We chose Finnish as our target language for this work, because it exemplifies many of the problems morphologically complex languages present for SMT. Among all the languages in the Europarl data-set, Finnish is the most difficult language to translate from and into, as was demonstrated in the MT Summit shared task (Koehn, 2005) . Another reason is the current lack of knowledge about how to apply SMT successfully to agglutinative languages like Turkish or Finnish. Our main contributions are: 1) the introduction of the notion of segmented translation where we explicitly allow phrase pairs that can end with a dangling morpheme, which can connect with other morphemes as part of the translation process, and 2) the use of a fully segmented translation model in combination with a post-processing morpheme prediction system, using unsupervised morphology induction. Both of these approaches beat the state of the art on the English-Finnish translation task. Morphology can express both content and function categories, and our experiments show that it is important to use morphology both within the translation model (for morphology with content) and outside it (for morphology contributing to fluency) . Automatic evaluation measures for MT, BLEU (Papineni et al., 2002), WER (Word Error Rate) and PER (Position Independent Word Error Rate) use the word as the basic unit rather than morphemes. In a word comProce dPinogrstla ofn tdh,e O 4r9etghon A,n Jnu nael 1 M9-e 2t4i,n2g 0 o1f1 t.he ?c A2s0s1o1ci Aatsiosonc fioartio Cno fmorpu Ctoamtiopnuatalt Lioin gauli Lsitnicgsu,i psatgices 32–42, prised of multiple morphemes, getting even a single morpheme wrong means the entire word is wrong. In addition to standard MT evaluation measures, we perform a detailed linguistic analysis of the output. Our proposed approaches are significantly better than the state of the art, achieving the highest reported BLEU scores on the English-Finnish Europarl version 3 data-set. Our linguistic analysis shows that our models have fewer morpho-syntactic errors compared to the word-based baseline. 2 2.1 Models Baseline Models We set up three baseline models for comparison in this work. The first is a basic wordbased model (called Baseline in the results) ; we trained this on the original unsegmented version of the text. Our second baseline is a factored translation model (Koehn and Hoang, 2007) (called Factored) , which used as factors the word, “stem” 1 and suffix. These are derived from the same unsupervised segmentation model used in other experiments. The results (Table 3) show that a factored model was unable to match the scores of a simple wordbased baseline. We hypothesize that this may be an inherently difficult representational form for a language with the degree of morphological complexity found in Finnish. Because the morphology generation must be precomputed, for languages with a high degree of morphological complexity, the combinatorial explosion makes it unmanageable to capture the full range of morphological productivity. In addition, because the morphological variants are generated on a per-word basis within a given phrase, it excludes productive morphological combination across phrase boundaries and makes it impossible for the model to take into account any longdistance dependencies between morphemes. We conclude from this result that it may be more useful for an agglutinative language to use morphology beyond the confines of the phrasal unit, and condition its generation on more than just the local target stem. In order to compare the 1see Section 2.2. 33 performance of unsupervised segmentation for translation, our third baseline is a segmented translation model based on a supervised segmentation model (called Sup) , using the hand-built Omorfi morphological analyzer (Pirinen and Listenmaa, 2007) , which provided slightly higher BLEU scores than the word-based baseline. 2.2 Segmented Translation For segmented translation models, it cannot be taken for granted that greater linguistic accuracy in segmentation yields improved translation (Chang et al. , 2008) . Rather, the goal in segmentation for translation is instead to maximize the amount of lexical content-carrying morphology, while generalizing over the information not helpful for improving the translation model. We therefore trained several different segmentation models, considering factors of granularity, coverage, and source-target symmetry. We performed unsupervised segmentation of the target data, using Morfessor (Creutz and Lagus, 2005) and Paramor (Monson, 2008) , two top systems from the Morpho Challenge 2008 (their combined output was the Morpho Challenge winner) . However, translation models based upon either Paramor alone or the combined systems output could not match the wordbased baseline, so we concentrated on Morfessor. Morfessor uses minimum description length criteria to train a HMM-based segmentation model. When tested against a human-annotated gold standard of linguistic morpheme segmentations for Finnish, this algorithm outperforms competing unsupervised methods, achieving an F-score of 67.0% on a 3 million sentence corpus (Creutz and Lagus, 2006) . Varying the perplexity threshold in Morfessor does not segment more word types, but rather over-segments the same word types. In order to get robust, common segmentations, we trained the segmenter on the 5000 most frequent words2 ; we then used this to segment the entire data set. In order to improve coverage, we then further segmented 2For the factored model baseline we also used the same setting perplexity = 30, 5,000 most frequent words, but with all but the last suffix collapsed and called the “stem” . TabHMleoat1nr:gplhiMngor phermphocTur631ae04in, 81c9ie03ns67gi,64n0S14e567theTp 2rsa51t, 29Se 3t168able and in translation. any word type that contained a match from the most frequent suffix set, looking for the longest matching suffix character string. We call this method Unsup L-match. After the segmentation, word-internal morpheme boundary markers were inserted into the segmented text to be used to reconstruct the surface forms in the MT output. We then trained the Moses phrase-based system (Koehn et al., 2007) on the segmented and marked text. After decoding, it was a simple matter to join together all adjacent morphemes with word-internal boundary markers to reconstruct the surface forms. Figure 1(a) gives the full model overview for all the variants of the segmented translation model (supervised/unsupervised; with and without the Unsup L-match procedure) . Table 1shows how morphemes are being used in the MT system. Of the phrases that included segmentations (‘Morph’ in Table 1) , roughly a third were ‘productive’, i.e. had a hanging morpheme (with a form such as stem+) that could be joined to a suffix (‘Hanging Morph’ in Table 1) . However, in phrases used while decoding the development and test data, roughly a quarter of the phrases that generated the translated output included segmentations, but of these, only a small fraction (6%) had a hanging morpheme; and while there are many possible reasons to account for this we were unable to find a single convincing cause. 2.3 Morphology Generation Morphology generation as a post-processing step allows major vocabulary reduction in the translation model, and allows the use of morphologically targeted features for modeling inflection. A possible disadvantage of this approach is that in this model there is no opportunity to con34 sider the morphology in translation since it is removed prior to training the translation model. Morphology generation models can use a variety of bilingual and contextual information to capture dependencies between morphemes, often more long-distance than what is possible using n-gram language models over morphemes in the segmented model. Similar to previous work (Minkov et al. , 2007; Toutanova et al. , 2008) , we model morphology generation as a sequence learning problem. Un- like previous work, we use unsupervised morphology induction and use automatically generated suffix classes as tags. The first phase of our morphology prediction model is to train a MT system that produces morphologically simplified word forms in the target language. The output word forms are complex stems (a stem and some suffixes) but still missing some important suffix morphemes. In the second phase, the output of the MT decoder is then tagged with a sequence of abstract suffix tags. In particular, the output of the MT decoder is a sequence of complex stems denoted by x and the output is a sequence of suffix class tags denoted by y. We use a list of parts from (x,y) and map to a d-dimensional feature vector Φ(x, y) , with each dimension being a real number. We infer the best sequence of tags using: F(x) = argymaxp(y | x,w) where F(x) returns the highest scoring output y∗ . A conditional random field (CRF) (Lafferty et al. , 2001) defines the conditional probability as a linear score for each candidate y and a global normalization term: logp(y | x, w) = Φ(x, y) · w − log Z where Z = Py0∈ exp(Φ(x, y0) · w) . We use stochastiPc gradient descent (using crfsgd3) to train the weight vector w. So far, this is all off-the-shelf sequence learning. However, the output y∗ from the CRF decoder is still only a sequence of abstract suffix tags. The third and final phase in our morphology prediction model GEN(x) 3 http://leon. bottou. org/projects/sgd English Training Data words Finnish Training Data words Morphological Pre-Processing stem+ +morph MT System Alignment: word word word stem+ +morph stem stem+ +morph Post-Process: Morph Re-Stitching Fully inflected surface form Evaluation against original reference (a) Segmented Translation Model English Training Data words Finnish Training Data Morphological Pre-Prowceosrdsisng 1 stem+ +morph1+ +morph2 Morphological Pre-Processing 2 stem+ +morph1+ MPosrpthe-mPRr+eo-+cSmetsio crhp1i:nhg+swteomrd+ MA+lTmigwnSomyrspdthen 1mt:+ wsotermd complex stem: stem+morph1+ MPo rpsht-oPlro gcyesGse2n:erCaRtioFnstem+morph1+ morph2sLuarnfagcueagfeorMmomdealp ing Fully inflected surface form Evaluation against original reference (b) Post-Processing Model Translation & Generation Figure 1: Training and testing pipelines for the SMT models. is to take the abstract suffix tag sequence y∗ and then map it into fully inflected word forms, and rank those outputs using a morphemic language model. The abstract suffix tags are extracted from the unsupervised morpheme learning process, and are carefully designed to enable CRF training and decoding. We call this model CRFLM for short. Figure 1(b) shows the full pipeline and Figure 2 shows a worked example of all the steps involved. We use the morphologically segmented training data (obtained using the segmented corpus described in Section 2.24) and remove selected suffixes to create a morphologically simplified version of the training data. The MT model is trained on the morphologically simplified training data. The output from the MT system is then used as input to the CRF model. The CRF model was trained on a ∼210,000 Finnish sentences, consisting noefd d∼ o1n.5 a am ∼il2li1o0n,0 tokens; tishhe 2,000 cseens,te cnoncse Europarl t.e5s tm isl eito nco tnoskiesntesd; hoef 41,434 stem tokens. The labels in the output sequence y were obtained by selecting the most productive 150 stems, and then collapsing certain vowels into equivalence classes corresponding to Finnish vowel harmony patterns. Thus 4Note that unlike Section 2.2 we do not use Unsup L-match because when evaluating the CRF model on the suffix prediction task it obtained 95.61% without using Unsup L-match and 82.99% when using Unsup L-match. 35 variants -k¨ o and -ko become vowel-generic enclitic particle -kO, and variants -ss ¨a and -ssa become the vowel-generic inessive case marker -ssA, etc. This is the only language-specific component of our translation model. However, we expect this approach to work for other agglutinative languages as well. For fusional languages like Spanish, another mapping from suffix to abstract tags might be needed. These suffix transformations to their equivalence classes prevent morphophonemic variants of the same morpheme from competing against each other in the prediction model. This resulted in 44 possible label outputs per stem which was a reasonable sized tag-set for CRF training. The CRF was trained on monolingual features of the segmented text for suffix prediction, where t is the current token: Word Stem st−n, .., st, .., st+n(n = 4) Morph Prediction yt−2 , yt−1 , yt With this simple feature set, we were able to use features over longer distances, resulting in a total of 1,110,075 model features. After CRF based recovery of the suffix tag sequence, we use a bigram language model trained on a full segmented version on the training data to recover the original vowels. We used bigrams only, because the suffix vowel harmony alternation depends only upon the preceding phonemes in the word from which it was segmented. original training koskevaa mietint o¨ ¨a data: k ¨asitell ¨a ¨an segmentation: koske+ +va+ +a mietint ¨o+ + a¨ k a¨si+ +te+ +ll a¨+ + a¨+ +n (train bigram language model with mapping A = { a , a }) map n fi bniaglr asmuff liaxn gtou agbest mraocdte tag-set: koske+ +va+ +A mietint ¨o+ +A k ¨asi+ +te+ +ll ¨a+ + ¨a+ +n (train CRF model to predict the final suffix) peeling of final suffix: koske+ +va+ mietint ¨o+ k a¨si+ +te+ +ll a¨+ + a¨+ (train SMT model on this transformation of training data) (a) Training decoder output: koske+ +va+ mietint o¨+ k a¨si+ +te+ +ll a¨+ + a¨+ decoder output stitched up: koskeva+ mietint o¨+ k ¨asitell ¨a ¨a+ CRF model prediction: x = ‘koskeva+ mietint ¨o+ k ¨asitell ¨a ¨a+’, y = ‘+A +A +n’ koskeva+ +A mietint ¨o+ +A k ¨asitell a¨ ¨a+ +n unstitch morphemes: koske+ +va+ +A mietint ¨o+ +A k ¨asi+ +te+ +ll ¨a+ + ¨a+ +n language model disambiguation: koske+ +va+ +a mietint ¨o+ + a¨ k a¨si+ +te+ +ll a¨+ + a¨+ +n final stitching: koskevaa mietint o¨ ¨a k ¨asitell ¨a ¨an (the output is then compared to the reference translation) (b) Decoding Figure 2: Worked example of all steps in the post-processing morphology prediction model. 3 Experimental Results used the Europarl version 3 corpus (Koehn, 2005) English-Finnish training data set, as well as the standard development and test data sets. Our parallel training data consists of ∼1 million senFor all of the models built in this paper, we tpeanrcaelsle lo tfr a4i0n nwgor ddast or less, sw ohfi ∼le 1t mhei development and test sets were each 2,000 sentences long. In all the experiments conducted in this paper, we used the Moses5 phrase-based translation system (Koehn et al. , 2007) , 2008 version. We trained all of the Moses systems herein using the standard features: language model, reordering model, translation model, and word penalty; in addition to these, the factored experiments called for additional translation and generation features for the added factors as noted above. We used in all experiments the following settings: a hypothesis stack size 100, distortion limit 6, phrase translations limit 20, and maximum phrase length 20. For the language models, we used SRILM 5-gram language models (Stolcke, 2002) for all factors. For our word-based Baseline system, we trained a word-based model using the same Moses system with identical settings. For evaluation against segmented translation systems in segmented forms before word reconstruction, we also segmented the baseline system’s word-based output. All the BLEU scores reported are for lowercase evaluation. We did an initial evaluation of the segmented output translation for each system using the no5http://www.statmt.org/moses/ 36 TabSlBUeuna2gps:meulSipengLmta-e nioatedchMo12dme804-.lB8S714cL±oEr0eUs.6 9 S8up19Nre.358ofe498rUs9ntoihe supervised segmentation baseline model. m-BLEU indicates that the segmented output was evaluated against a segmented version of the reference (this measure does not have the same correlation with human judgement as BLEU) . No Uni indicates the segmented BLEU score without unigrams. tion of m-BLEU score (Luong et al. , 2010) where the BLEU score is computed by comparing the segmented output with a segmented reference translation. Table 2 shows the m-BLEU scores for various systems. We also show the m-BLEU score without unigrams, since over-segmentation could lead to artificially high m-BLEU scores. In fact, if we compare the relative improvement of our m-BLEU scores for the Unsup L-match system we see a relative improvement of 39.75% over the baseline. Luong et. al. (2010) report an m-BLEU score of 55.64% but obtain a relative improvement of 0.6% over their baseline m-BLEU score. We find that when using a good segmentation model, segmentation of the morphologically complex target language improves model performance over an unsegmented baseline (the confidence scores come from bootstrap resampling) . Table 3 shows the evaluation scores for all the baselines and the methods introduced in this paper using standard wordbased lowercase BLEU, WER and PER. We do TSCMaFBU(LubanRolpcesdFotu3lne-ipLr:gMdeLT-tms.al,Stc2ho0r1es:)l 1wB54 Le.r682E90c 27a9Us∗eBL-7 W46E3. U659478R6,1WE-7 TR412E. 847Ra1528nd TER. The ∗ indicates a statistically significant improvement o∗f BndLiEcaUte score over tchalel yB saisgenli nfice mntod imel.The boldface scores are the best performing scores per evaluation measure. better than (Luong et al. , 2010) , the previous best score for this task. We also show a better relative improvement over our baseline when compared to (Luong et al., 2010) : a relative improvement of 4.86% for Unsup L-match compared to our baseline word-based model, compared to their 1.65% improvement over their baseline word-based model. Our best performing method used unsupervised morphology with L-match (see Section 2.2) and the improvement is significant: bootstrap resampling provides a confidence margin of ±0.77 and a t-test (Collins ceot nafli.d , 2005) sahrogwined o significance aw ti-thte p = 0o.0ll0in1s. 3.1 Morphological Fluency Analysis To see how well the models were doing at getting morphology right, we examined several patterns of morphological behavior. While we wish to explore minimally supervised morphological MT models, and use as little language specific information as possible, we do want to use linguistic analysis on the output of our system to see how well the models capture essential morphological information in the target language. So, we ran the word-based baseline system, the segmented model (Unsup L-match) , and the prediction model (CRF-LM) outputs, along with the reference translation through the supervised morphological analyzer Omorfi (Pirinen and Listenmaa, 2007) . Using this analysis, we looked at a variety of linguistic constructions that might reveal patterns in morphological behavior. These were: (a) explicitly marked 37 noun forms, (b) noun-adjective case agreement, (c) subject-verb person/number agreement, (d) transitive object case marking, (e) postpositions, and (f) possession. In each of these categories, we looked for construction matches on a per-sentence level between the models’ output and the reference translation. Table 4 shows the models’ performance on the constructions we examined. In all of the categories, the CRF-LM model achieves the best precision score, as we explain below, while the Unsup L-match model most frequently gets the highest recall score. A general pattern in the most prevalent of these constructions is that the baseline tends to prefer the least marked form for noun cases (corresponding to the nominative) more than the reference or the CRF-LM model. The baseline leaves nouns in the (unmarked) nominative far more than the reference, while the CRF-LM model comes much closer, so it seems to fare better at explicitly marking forms, rather than defaulting to the more frequent unmarked form. Finnish adjectives must be marked with the same case as their head noun, while verbs must agree in person and number with their subject. We saw that in both these categories, the CRFLM model outperforms for precision, while the segmented model gets the best recall. In addition, Finnish generally marks direct objects of verbs with the accusative or the partitive case; we observed more accusative/partitive-marked nouns following verbs in the CRF-LM output than in the baseline, as illustrated by example (1) in Fig. 3. While neither translation picks the same verb as in the reference for the input ‘clarify,’ the CRFLM-output paraphrases it by using a grammatical construction of the transitive verb followed by a noun phrase inflected with the accusative case, correctly capturing the transitive construction. The baseline translation instead follows ‘give’ with a direct object in the nominative case. To help clarify the constructions in question, we have used Google Translate6 to provide back6 http://translate.google. com/ of occurrences per sentence, recall and F-score. also averaged The constructions over the various translations. are listed in descending P, R and F stand for precision, order of their frequency in the texts. The highlighted value in each column is the most accurate with respect to the reference value. translations of our MT output into English; to contextualize these back-translations, we have provided Google’s back-translation of the reference. The use of postpositions shows another difference between the models. Finnish postpositions require the preceding noun to be in the genitive or sometimes partitive case, which occurs correctly more frequently in the CRF-LM than the baseline. In example (2) in Fig. 3, all three translations correspond to the English text, ‘with the basque nationalists. ’ However, the CRF-LM output is more grammatical than the baseline, because not only do the adjective and noun agree for case, but the noun ‘baskien’ to which the postposition ‘kanssa’ belongs is marked with the correct genitive case. However, this well-formedness is not rewarded by BLEU, because ‘baskien’ does not match the reference. In addition, while Finnish may express possession using case marking alone, it has another construction for possession; this can disambiguate an otherwise ambiguous clause. This alternate construction uses a pronoun in the genitive case followed by a possessive-marked noun; we see that the CRF-LM model correctly marks this construction more frequently than the baseline. As example (3) in Fig. 3 shows, while neither model correctly translates ‘matkan’ (‘trip’) , the baseline’s output attributes the inessive ‘yhteydess’ (‘connection’) as belonging to ‘tulokset’ (‘results’) , and misses marking the possession linking it to ‘Commissioner Fischler’. Our manual evaluation shows that the CRF38 LM model is producing output translations that are more morphologically fluent than the wordbased baseline and the segmented translation Unsup L-match system, even though the word choices lead to a lower BLEU score overall when compared to Unsup L-match. 4 Related Work The work on morphology in MT can be grouped into three categories, factored models, segmented translation, and morphology generation. Factored models (Koehn and Hoang, 2007) factor the phrase translation probabilities over additional information annotated to each word, allowing for text to be represented on multiple levels of analysis. We discussed the drawbacks of factored models for our task in Section 2. 1. While (Koehn and Hoang, 2007; Yang and Kirchhoff, 2006; Avramidis and Koehn, 2008) obtain improvements using factored models for translation into English, German, Spanish, and Czech, these models may be less useful for capturing long-distance dependencies in languages with much more complex morphological systems such as Finnish. In our experiments factored models did worse than the baseline. Segmented translation performs morphological analysis on the morphologically complex text for use in the translation model (Brown et al. , 1993; Goldwater and McClosky, 2005; de Gispert and Mari n˜o, 2008) . This method unpacks complex forms into simpler, more frequently occurring components, and may also increase the symmetry of the lexically realized content be(1) Input: ‘the charter we are to approve today both strengthens and gives visible shape to the common fundamental rights and values our community is to be based upon. ’ a. Reference: perusoikeuskirja , jonka t ¨an ¨a ¨an aiomme hyv a¨ksy ¨a , sek ¨a vahvistaa ett ¨a selvent a¨ a¨ (selvent ¨a a¨/VERB/ACT/INF/SG/LAT-clarify) niit a¨ (ne/PRONOUN/PL/PAR-them) yhteisi ¨a perusoikeuksia ja arvoja , joiden on oltava yhteis¨ omme perusta. Back-translation: ‘Charter of Fundamental Rights, which today we are going to accept that clarify and strengthen the common fundamental rights and values, which must be community based. ’ b. Baseline: perusoikeuskirja me hyv ¨aksymme t¨ an ¨a a¨n molemmat vahvistaa ja antaa (antaa/VERB/INF/SG/LATgive) n a¨kyv a¨ (n¨ aky a¨/VERB/ACT/PCP/SG/NOM-visible) muokata yhteist ¨a perusoikeuksia ja arvoja on perustuttava. Back-translation: ‘Charter today, we accept both confirm and modify to make a visible and common values, fundamental rights must be based. ’ c. CRF-LM: perusoikeuskirja on hyv a¨ksytty t ¨an ¨a ¨an , sek ¨a vahvistaa ja antaa (antaa/VERB/ACT/INF/SG/LAT-give) konkreettisen (konkreettinen/ADJECTIVE/SG/GEN,ACC-concrete) muodon (muoto/NOUN/SG/GEN,ACCshape) yhteisi ¨a perusoikeuksia ja perusarvoja , yhteis¨ on on perustuttava. Back-translation: ‘Charter has been approved today, and to strengthen and give concrete shape to the common basic rights and fundamental values, the Community must be based. ’ (2) Input: ‘with the basque nationalists’ a. Reference: baskimaan kansallismielisten kanssa basque-SG/NOM+land-SG/GEN,ACC nationalists-PL/GEN with-POST b. Baseline: baskimaan kansallismieliset kanssa basque-SG/NOM-+land-SG/GEN,ACC kansallismielinen-PL/NOM,ACC-nationalists POST-with c. CRF-LM: kansallismielisten baskien kanssa nationalists-PL/GEN basques-PL/GEN with-POST (3) Input: ‘and in this respect we should value the latest measures from commissioner fischler , the results of his trip to morocco on the 26th of last month and the high level meetings that took place, including the one with the king himself’ a. Reference: ja t ¨ass¨ a mieless ¨a osaamme my¨ os arvostaa komission j¨ asen fischlerin viimeisimpi ¨a toimia , jotka ovat h a¨nen (h¨ anen/GEN-his) marokkoon 26 lokakuuta tekemns (tekem¨ ans ¨a/POSS-his) matkan (matkan/GENtour) ja korkean tason kokousten jopa itsens¨ a kuninkaan kanssa tulosta Back-translation: ‘and in this sense we can also appreciate the Commissioner Fischler’s latest actions, which are his to Morocco 26 October trip to high-level meetings and even the king himself with the result b. Baseline: ja t ¨ass¨ a yhteydess a¨ olisi arvoa viimeisin toimia komission j¨ asen fischler , tulokset monitulkintaisia marokon yhteydess a¨ (yhteydess/INE-connection) , ja viime kuussa pidettiin korkean tason kokouksissa , mukaan luettuna kuninkaan kanssa Back-translation: ‘and in this context would be the value of the last act, Commissioner Fischler, the results of the Moroccan context, ambiguous, and last month held high level meetings, including with the king’ c. CRF-LM: ja t ¨ass¨ a yhteydess a¨ meid ¨an olisi lis ¨aarvoa viimeist ¨a toimenpiteit a¨ kuin komission j¨ asen fischler , ett a¨ h a¨nen (h¨ anen/GEN-his) kokemuksensa (kokemuksensa/POSS-experience) marokolle (marokolle-Moroccan) viime kuun 26 ja korkean tason tapaamiset j¨ arjestettiin, kuninkaan kanssa Back-translation: ‘and in this context, we should value the last measures as the Commissioner Fischler, that his experience in Morocco has on the 26th and high-level meetings took place, including with the king. ’ Figure 3: Morphological fluency analysis (see Section 3. 1) . tween source and target. In a somewhat orthogonal approach to ours, (Ma et al. , 2007) use alignment of a parallel text to pack together adjacent segments in the alignment output, which are then fed back to the word aligner to bootstrap an improved alignment, which is then used in the translation model. We compared our results against (Luong et al. , 2010) in Table 3 since their results are directly comparable to ours. They use a segmented phrase table and language model along with the word-based versions in the decoder and in tuning a Finnish target. Their approach requires segmented phrases 39 to match word boundaries, eliminating morphologically productive phrases. In their work a segmented language model can score a translation, but cannot insert morphology that does not show source-side reflexes. In order to perform a similar experiment that still allowed for morphologically productive phrases, we tried training a segmented translation model, the output of which we stitched up in tuning so as to tune to a word-based reference. The goal of this experiment was to control the segmented model’s tendency to overfit by rewarding it for using correct whole-word forms. However, we found that this approach was less successful than using the segmented reference in tuning, and could not meet the baseline (13.97% BLEU best tuning score, versus 14.93% BLEU for the baseline best tuning score) . Previous work in segmented translation has often used linguistically motivated morphological analysis selectively applied based on a language-specific heuristic. A typical approach is to select a highly inflecting class of words and segment them for particular morphology (de Gispert and Mari n˜o, 2008; Ramanathan et al. , 2009) . Popovi¸ c and Ney (2004) perform segmentation to reduce morphological complexity of the source to translate into an isolating target, reducing the translation error rate for the English target. For Czech-to-English, Goldwater and McClosky (2005) lemmatized the source text and inserted a set of ‘pseudowords’ expected to have lexical reflexes in English. Minkov et. al. (2007) and Toutanova et. al. (2008) use a Maximum Entropy Markov Model for morphology generation. The main drawback to this approach is that it removes morphological information from the translation model (which only uses stems) ; this can be a problem for languages in which morphology ex- presses lexical content. de Gispert (2008) uses a language-specific targeted morphological classifier for Spanish verbs to avoid this issue. Talbot and Osborne (2006) use clustering to group morphological variants of words for word alignments and for smoothing phrase translation tables. Habash (2007) provides various methods to incorporate morphological variants of words in the phrase table in order to help recognize out of vocabulary words in the source language. 5 Conclusion and Future Work We found that using a segmented translation model based on unsupervised morphology induction and a model that combined morpheme segments in the translation model with a postprocessing morphology prediction model gave us better BLEU scores than a word-based baseline. Using our proposed approach we obtain better scores than the state of the art on the EnglishFinnish translation task (Luong et al. , 2010) : from 14.82% BLEU to 15.09%, while using a 40 simpler model. We show that using morphological segmentation in the translation model can improve output translation scores. We also demonstrate that for Finnish (and possibly other agglutinative languages) , phrase-based MT benefits from allowing the translation model access to morphological segmentation yielding productive morphological phrases. Taking advantage of linguistic analysis of the output we show that using a post-processing morphology generation model can improve translation fluency on a sub-word level, in a manner that is not captured by the BLEU word-based evaluation measure. In order to help with replication of the results in this paper, we have run the various morphological analysis steps and created the necessary training, tuning and test data files needed in order to train, tune and test any phrase-based machine translation system with our data. The files can be downloaded from natlang. cs.sfu. ca. In future work we hope to explore the utility of phrases with productive morpheme boundaries and explore why they are not used more pervasively in the decoder. Evaluation measures for morphologically complex languages and tun- ing to those measures are also important future work directions. Also, we would like to explore a non-pipelined approach to morphological preand post-processing so that a globally trained model could be used to remove the target side morphemes that would improve the translation model and then predict those morphemes in the target language. Acknowledgements This research was partially supported by NSERC, Canada (RGPIN: 264905) and a Google Faculty Award. We would like to thank Christian Monson, Franz Och, Fred Popowich, Howard Johnson, Majid Razmara, Baskaran Sankaran and the anonymous reviewers for their valuable comments on this work. We would particularly like to thank the developers of the open-source Moses machine translation toolkit and the Omorfi morphological analyzer for Finnish which we used for our experiments. References Eleftherios Avramidis and Philipp Koehn. 2008. Enriching morphologically poor languages for statistical machine translation. In Proceedings of the 46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, page 763?770, Columbus, Ohio, USA. Association for Computational Linguistics. Peter F. Brown, Stephen A. Della Pietra, Vincent J. Della Pietra, and R. L. Mercer. 1993. The mathematics of statistical machine translation: Parameter estimation. Computational Linguistics, 19(2) :263–31 1. Pi-Chuan Chang, Michel Galley, and Christopher D. Manning. 2008. Optimizing Chinese word segmentation for machine translation performance. In Proceedings of the Third Workshop on Statistical Machine Translation, pages 224–232, Columbus, Ohio, June. Association for Computational Linguistics. Michael Collins, Philipp Koehn, and Ivona Kucerova. 2005. Clause restructuring for statistical machine translation. In Proceedings of 43rd Annual Meeting of the Association for Computational Linguistics (A CL05). Association for Computational Linguistics. Mathias Creutz and Krista Lagus. 2005. Inducing the morphological lexicon of a natural language from unannotated text. In Proceedings of the International and Interdisciplinary Conference on Adaptive Knowledge Representation and Reason- ing (AKRR ’05), pages 106–113, Espoo, Finland. Mathias Creutz and Krista Lagus. 2006. Morfessor in the morpho challenge. In Proceedings of the PASCAL Challenge Workshop on Unsupervised Segmentation of Words into Morphemes. Adri ´a de Gispert and Jos e´ Mari n˜o. 2008. On the impact of morphology in English to Spanish statistical MT. Speech Communication, 50(11-12) . Sharon Goldwater and David McClosky. 2005. Improving statistical MT through morphological analysis. In Proceedings of the Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, pages 676–683, Vancouver, B.C. , Canada. Association for Computational Linguistics. Philipp Koehn and Hieu Hoang. 2007. Factored translation models. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 868–876, Prague, Czech Republic. Association for Computational Linguistics. 41 Philipp Koehn, Hieu Hoang, Alexandra Birch, Chris Callison-Burch, Marcello Federico, Nicola Bertoldi, Brooke Cowan, Wade Shen, Christine Moran, Richard Zens, Chris Dyer, Ondrej Bojar, Alexandra Constantin, and Evan Herbst. 2007. Moses: Open source toolkit for statistical machine translation. In A CL ‘07: Proceedings of the 45th Annual Meeting of the A CL on Interactive Poster and Demonstration Sessions, pages 177–108, Prague, Czech Republic. Association for Computational Linguistics. Philipp Koehn. 2005. Europarl: A parallel corpus for statistical machine translation. In Proceedings of Machine Translation Summit X, pages 79–86, Phuket, Thailand. Association for Computational Linguistics. John Lafferty, Andrew McCallum, and Fernando Pereira. 2001. Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In Proceedings of the 18th International Conference on Machine Learning, pages 282–289, San Francisco, California, USA. Association for Computing Machinery. Minh-Thang Luong, Preslav Nakov, and Min-Yen Kan. 2010. A hybrid morpheme-word representation for machine translation of morphologically rich languages. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 148–157, Cambridge, Massachusetts. Association for Computational Linguistics. Yanjun Ma, Nicolas Stroppa, and Andy Way. 2007. Bootstrapping word alignment via word packing. In Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics, pages 304–311, Prague, Czech Republic. Association for Computational Linguistics. Einat Minkov, Kristina Toutanova, and Hisami Suzuki. 2007. Generating complex morphology for machine translation. In In Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics (A CL07), pages 128–135, Prague, Czech Republic. Association for Computational Linguistics. Christian Monson. 2008. Paramor and morpho challenge 2008. In Lecture Notes in Computer Science: Workshop of the Cross-Language Evaluation Forum (CLEF 2008), Revised Selected Papers. Habash Nizar. 2007. Four techniques for online handling of out-of-vocabulary words in arabic-english statistical machine translation. In Proceedings of the 46th Annual Meeting of the Association of Computational Linguistics, Columbus, Ohio. Association for Computational Linguistics. Kishore Papineni, Salim Roukos, Todd Ward, and Zhu. 2002. BLEU: A method for automatic evaluation of machine translation. In Proceedings of 40th Annual Meeting of the Association for Computational Linguistics A CL, pages 311–318, Philadelphia, Pennsylvania, USA. Association for Computational Linguistics. Tommi Pirinen and Inari Listenmaa. 2007. Omorfi morphological analzer. http://gna.org/projects/omorfi. Maja Popovi¸ c and Hermann Ney. 2004. Towards the use of word stems and suffixes for statistiWei jing cal machine translation. In Proceedings of the 4th International Conference on Language Resources and Evaluation (LREC), pages 1585–1588, Lisbon, Portugal. European Language Resources Association (ELRA) . Ananthakrishnan Ramanathan, Hansraj Choudhary, Avishek Ghosh, and Pushpak Bhattacharyya. 2009. Case markers and morphology: Addressing the crux of the fluency problem in EnglishHindi SMT. In Proceedings of the Joint Conference of the 4 7th Annual Meeting of the Association for Computational Linguistics and the 4th International Joint Conference on Natural Language Processing of the Asian Federation of Natural Language Processing, pages 800–808, Suntec, Singapore. Association for Computational Linguistics. Andreas Stolcke. 2002. Srilm – an extensible language modeling toolkit. 7th International Conference on Spoken Language Processing, 3:901–904. David Talbot and Miles Osborne. 2006. Modelling lexical redundancy for machine translation. In Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the Association for Computational Linguistics, pages 969–976, Sydney, Australia, July. Association for Computational Linguistics. Kristina Toutanova, Hisami Suzuki, and Achim Ruopp. 2008. Applying morphology generation models to machine translation. In Proceedings of the 46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, pages 514–522, Columbus, Ohio, USA. Association for Computational Linguistics. Mei Yang and Katrin Kirchhoff. 2006. Phrase-based backoff models for machine translation of highly inflected languages. In Proceedings of the European Chapter of the Association for Computational Linguistics, pages 41–48, Trento, Italy. Association for Computational Linguistics. 42
5 0.58959502 10 acl-2011-A Discriminative Model for Joint Morphological Disambiguation and Dependency Parsing
Author: John Lee ; Jason Naradowsky ; David A. Smith
Abstract: Most previous studies of morphological disambiguation and dependency parsing have been pursued independently. Morphological taggers operate on n-grams and do not take into account syntactic relations; parsers use the “pipeline” approach, assuming that morphological information has been separately obtained. However, in morphologically-rich languages, there is often considerable interaction between morphology and syntax, such that neither can be disambiguated without the other. In this paper, we propose a discriminative model that jointly infers morphological properties and syntactic structures. In evaluations on various highly-inflected languages, this joint model outperforms both a baseline tagger in morphological disambiguation, and a pipeline parser in head selection.
6 0.57780111 124 acl-2011-Exploiting Morphology in Turkish Named Entity Recognition System
7 0.55434954 310 acl-2011-Translating from Morphologically Complex Languages: A Paraphrase-Based Approach
8 0.50678021 303 acl-2011-Tier-based Strictly Local Constraints for Phonology
9 0.50554746 151 acl-2011-Hindi to Punjabi Machine Translation System
10 0.43585995 311 acl-2011-Translationese and Its Dialects
11 0.42350778 157 acl-2011-I Thou Thee, Thou Traitor: Predicting Formal vs. Informal Address in English Literature
12 0.41720784 16 acl-2011-A Joint Sequence Translation Model with Integrated Reordering
13 0.41201654 163 acl-2011-Improved Modeling of Out-Of-Vocabulary Words Using Morphological Classes
14 0.40275297 62 acl-2011-Blast: A Tool for Error Analysis of Machine Translation Output
15 0.39005786 44 acl-2011-An exponential translation model for target language morphology
16 0.38443986 184 acl-2011-Joint Hebrew Segmentation and Parsing using a PCFGLA Lattice Parser
17 0.36838326 69 acl-2011-Clause Restructuring For SMT Not Absolutely Helpful
18 0.35935479 238 acl-2011-P11-2093 k2opt.pdf
19 0.35911 215 acl-2011-MACAON An NLP Tool Suite for Processing Word Lattices
20 0.35445821 6 acl-2011-A Comprehensive Dictionary of Multiword Expressions
topicId topicWeight
[(5, 0.029), (10, 0.014), (17, 0.065), (26, 0.055), (37, 0.075), (39, 0.071), (41, 0.07), (53, 0.012), (55, 0.039), (59, 0.04), (72, 0.019), (74, 0.215), (91, 0.035), (96, 0.146), (97, 0.022)]
simIndex simValue paperId paperTitle
same-paper 1 0.80672145 193 acl-2011-Language-independent compound splitting with morphological operations
Author: Klaus Macherey ; Andrew Dai ; David Talbot ; Ashok Popat ; Franz Och
Abstract: Translating compounds is an important problem in machine translation. Since many compounds have not been observed during training, they pose a challenge for translation systems. Previous decompounding methods have often been restricted to a small set of languages as they cannot deal with more complex compound forming processes. We present a novel and unsupervised method to learn the compound parts and morphological operations needed to split compounds into their compound parts. The method uses a bilingual corpus to learn the morphological operations required to split a compound into its parts. Furthermore, monolingual corpora are used to learn and filter the set of compound part candidates. We evaluate our method within a machine translation task and show significant improvements for various languages to show the versatility of the approach.
2 0.6944797 137 acl-2011-Fine-Grained Class Label Markup of Search Queries
Author: Joseph Reisinger ; Marius Pasca
Abstract: We develop a novel approach to the semantic analysis of short text segments and demonstrate its utility on a large corpus of Web search queries. Extracting meaning from short text segments is difficult as there is little semantic redundancy between terms; hence methods based on shallow semantic analysis may fail to accurately estimate meaning. Furthermore search queries lack explicit syntax often used to determine intent in question answering. In this paper we propose a hybrid model of semantic analysis combining explicit class-label extraction with a latent class PCFG. This class-label correlation (CLC) model admits a robust parallel approximation, allowing it to scale to large amounts of query data. We demonstrate its performance in terms of (1) its predicted label accuracy on polysemous queries and (2) its ability to accurately chunk queries into base constituents.
3 0.68947148 123 acl-2011-Exact Decoding of Syntactic Translation Models through Lagrangian Relaxation
Author: Alexander M. Rush ; Michael Collins
Abstract: We describe an exact decoding algorithm for syntax-based statistical translation. The approach uses Lagrangian relaxation to decompose the decoding problem into tractable subproblems, thereby avoiding exhaustive dynamic programming. The method recovers exact solutions, with certificates of optimality, on over 97% of test examples; it has comparable speed to state-of-the-art decoders.
4 0.68575013 15 acl-2011-A Hierarchical Pitman-Yor Process HMM for Unsupervised Part of Speech Induction
Author: Phil Blunsom ; Trevor Cohn
Abstract: In this work we address the problem of unsupervised part-of-speech induction by bringing together several strands of research into a single model. We develop a novel hidden Markov model incorporating sophisticated smoothing using a hierarchical Pitman-Yor processes prior, providing an elegant and principled means of incorporating lexical characteristics. Central to our approach is a new type-based sampling algorithm for hierarchical Pitman-Yor models in which we track fractional table counts. In an empirical evaluation we show that our model consistently out-performs the current state-of-the-art across 10 languages.
5 0.68465531 327 acl-2011-Using Bilingual Parallel Corpora for Cross-Lingual Textual Entailment
Author: Yashar Mehdad ; Matteo Negri ; Marcello Federico
Abstract: This paper explores the use of bilingual parallel corpora as a source of lexical knowledge for cross-lingual textual entailment. We claim that, in spite of the inherent difficulties of the task, phrase tables extracted from parallel data allow to capture both lexical relations between single words, and contextual information useful for inference. We experiment with a phrasal matching method in order to: i) build a system portable across languages, and ii) evaluate the contribution of lexical knowledge in isolation, without interaction with other inference mechanisms. Results achieved on an English-Spanish corpus obtained from the RTE3 dataset support our claim, with an overall accuracy above average scores reported by RTE participants on monolingual data. Finally, we show that using parallel corpora to extract paraphrase tables reveals their potential also in the monolingual setting, improving the results achieved with other sources of lexical knowledge.
6 0.68090653 178 acl-2011-Interactive Topic Modeling
7 0.68073654 324 acl-2011-Unsupervised Semantic Role Induction via Split-Merge Clustering
8 0.67986441 182 acl-2011-Joint Annotation of Search Queries
9 0.67975837 202 acl-2011-Learning Hierarchical Translation Structure with Linguistic Annotations
10 0.67896831 269 acl-2011-Scaling up Automatic Cross-Lingual Semantic Role Annotation
11 0.67799026 258 acl-2011-Ranking Class Labels Using Query Sessions
12 0.67760742 11 acl-2011-A Fast and Accurate Method for Approximate String Search
13 0.67708045 241 acl-2011-Parsing the Internal Structure of Words: A New Paradigm for Chinese Word Segmentation
14 0.67690718 58 acl-2011-Beam-Width Prediction for Efficient Context-Free Parsing
15 0.67513311 300 acl-2011-The Surprising Variance in Shortest-Derivation Parsing
16 0.67378521 209 acl-2011-Lexically-Triggered Hidden Markov Models for Clinical Document Coding
17 0.67333806 36 acl-2011-An Efficient Indexer for Large N-Gram Corpora
18 0.67309892 3 acl-2011-A Bayesian Model for Unsupervised Semantic Parsing
19 0.67274034 28 acl-2011-A Statistical Tree Annotator and Its Applications
20 0.67261183 126 acl-2011-Exploiting Syntactico-Semantic Structures for Relation Extraction