acl acl2013 acl2013-299 knowledge-graph by maker-knowledge-mining

299 acl-2013-Reconstructing an Indo-European Family Tree from Non-native English Texts


Source: pdf

Author: Ryo Nagata ; Edward Whittaker

Abstract: Mother tongue interference is the phenomenon where linguistic systems of a mother tongue are transferred to another language. Although there has been plenty of work on mother tongue interference, very little is known about how strongly it is transferred to another language and about what relation there is across mother tongues. To address these questions, this paper explores and visualizes mother tongue interference preserved in English texts written by Indo-European language speakers. This paper further explores linguistic features that explain why certain relations are preserved in English writing, and which contribute to related tasks such as native language identification.

Reference: text


Summary: the most important sentenses genereted by tfidf model

sentIndex sentText sentNum sentScore

1 Abstract Mother tongue interference is the phenomenon where linguistic systems of a mother tongue are transferred to another language. [sent-3, score-1.232]

2 Although there has been plenty of work on mother tongue interference, very little is known about how strongly it is transferred to another language and about what relation there is across mother tongues. [sent-4, score-1.04]

3 To address these questions, this paper explores and visualizes mother tongue interference preserved in English texts written by Indo-European language speakers. [sent-5, score-0.94]

4 This paper further explores linguistic features that explain why certain relations are preserved in English writing, and which contribute to related tasks such as native language identification. [sent-6, score-0.253]

5 1 Introduction Transfer of linguistic systems of a mother tongue to another language, namely mother tongue interference, is often observable in the writing of nonnative speakers. [sent-7, score-1.295]

6 The reader may be able to determine the mother tongue of the writer of the following sentence from the underlined article error: The alien wouldn ’t use my spaceship but the hers. [sent-8, score-0.71]

7 The answer would probably be French or Spanish; the definite article is allowed to modify possessive pronouns in these languages, and the usage is sometimes negatively transferred to English writing. [sent-9, score-0.229]

8 Researchers such as Swan and Smith (2001), Aarts and Granger (1998), DavidsenNielsen and Harder (2001), and Altenberg and Tapper (1998) work on mother tongue interference to reveal overused/underused words, part of speech (POS), or grammatical items. [sent-10, score-0.824]

9 uk , In contrast, very little is known about how strongly mother tongue interference is transferred to another language and about what relation there is across mother tongues. [sent-13, score-1.242]

10 At one extreme, one could argue that it is so strongly transferred to texts in another language that the linguistic relations between mother tongues are perfectly preserved in the texts. [sent-14, score-0.578]

11 At the other extreme, one can counter it, arguing that other features such as non-nativeness are more influential than mother tongue interference. [sent-15, score-0.622]

12 One possible reason for this is that a large part of the distinctive language systems of a mother tongue may be eliminated when transferred to another language from a speaker’s mother tongue. [sent-16, score-1.008]

13 However, the difference in the richness cannot be transferred into English because English has almost no inflectional case system. [sent-20, score-0.125]

14 Thus, one cannot determine the mother tongue of a given nonnative text from the inflectional case. [sent-21, score-0.712]

15 Besides, Wong and Dras (2009) show that there are no significant differences, between mother tongues, in the misuse of certain syntactic features such as subject-verb agreement that have different tendencies depending on their mother tongues. [sent-23, score-0.6]

16 We hypothesize that: Hypothesis: Mother tongue interference is so strong that the relations in a language family are preserved in texts written in another language. [sent-27, score-0.871]

17 In other words, mother tongue interference is so strong that one can reconstruct a language fam1137 Proce dingsS o f ita h,e B 5u1lgsta Arinan,u Aaulg Musete 4ti-n9g 2 o0f1 t3h. [sent-28, score-0.891]

18 One of the major contributions of this work is to reveal and visualize a language family tree preserved in non-native texts, by examining the hypothesis. [sent-31, score-0.375]

19 This becomes important in native language identification1 which is useful for improving grammatical error correction systems (Chodorow et al. [sent-32, score-0.195]

20 6, this paper reveals several crucial findings that contribute to improving native language identification. [sent-35, score-0.195]

21 In addition, this paper shows that the findings could contribute to reconstruction of language family trees (Enright and Kondrak, 2011; Gray and Atkinson, 2003; Barbanc ¸on et al. [sent-36, score-0.263]

22 2 Approach To examine the hypothesis, we reconstruct a language family tree from English texts written by non-native speakers of English whose mother tongue is one of the Indo-European languages (Beekes, 2011; Ramat and Ramat, 2006). [sent-51, score-1.151]

23 If the reconstructed tree is sufficiently similar to the original Indo-European family tree, it will support the hypothesis. [sent-52, score-0.377]

24 If not, it suggests that some features other than mother tongue interference are more influential. [sent-53, score-0.824]

25 The approach we use for reconstructing a language family tree is to apply agglomerative hierarchical clustering (Han and Kamber, 2006) to English texts written by non-native speakers. [sent-54, score-0.549]

26 Researchers have already performed related work on reconstructing language family trees. [sent-55, score-0.319]

27 (1992) and Kita (1999) proposed methods for reconstructing language family trees using clustering. [sent-58, score-0.351]

28 Among them, the 1Recently, native language identification has drawn the at- tention of NLP researchers. [sent-59, score-0.195]

29 For instance, a shared task on native language identification took place at an NAACL-HLT 2013 workshop. [sent-60, score-0.195]

30 Then, agglomerative hierarchical clustering is applied to the language models to reconstruct a language family tree. [sent-65, score-0.384]

31 The similarity used for clustering is based on a divergence-like distance between two language models that was originally proposed by Juang and Rabiner (1985). [sent-66, score-0.097]

32 Kita’s work (and other previous work) targets clustering of a variety of languages whereas our work tries to reconstruct a language family tree preserved in non-native English. [sent-70, score-0.534]

33 Obviously, this does not work on our task; belly is belly in English writing whoever writes it. [sent-74, score-0.098]

34 Although spelling is sometimes influenced by mother tongues, it involves a lot more including overuse, underuse, and misuse of lexical, grammatical, and syntactic systems. [sent-78, score-0.3]

35 To solve the problem, this work adopts a wordbased language model in the expectation that word sequences reflect mother tongue interference. [sent-79, score-0.622]

36 It would reflect the topics of given texts rather than mother tongue interference. [sent-81, score-0.68]

37 Let Di be a set of English 1138 texts where idenotes a mother tongue i. [sent-87, score-0.68]

38 EOS Note that the content of the original sentence is far from clear while reflecting mother tongue interference, especially in the hers. [sent-104, score-0.622]

39 The clustering algorithm used is agglomerative hierarchical clustering with the average linkage method. [sent-110, score-0.141]

40 To sum up, the procedure of the language fam- ily tree construction method is as follows: (i) Preprocess each Di; (ii) Build Mi from Di; (iii) Calculate the distances between the language models; (iv) Cluster the language data using the distances; (v) Output the result as a language family tree. [sent-120, score-0.348]

41 2 Vector-based Method We also examine a vector-based method for language family tree reconstruction. [sent-122, score-0.317]

42 5, this method allows us to interpret clustering results more easily than with the language model-based method while both result in similar language family trees. [sent-124, score-0.286]

43 The clustering procedure is the same as for the language model-based method except that Mi is vector-based and that the distance metric is Euclidean. [sent-128, score-0.097]

44 It consists of English essays written by a wide variety of nonnative speakers of English. [sent-132, score-0.201]

45 For reference, we also used native English (British and American university students’ essays in the LOCNESS corpus5) and two sets of Japanese English (ICLE and the NICE corpus (Sugiura et al. [sent-139, score-0.295]

46 Existing POS taggers might not perform well on non-native English texts because they are normally developed to analyze native English texts. [sent-143, score-0.253]

47 Although it did not perform as well as on native texts, it still achieved a fair accuracy. [sent-149, score-0.195]

48 3For example, because of (iii), essays written by native speakers of Swedish in the Finnish subcorpus were excluded from the experiments. [sent-153, score-0.399]

49 5The LOCNESS corpus is a corpus of native English essays made up of British pupils’ essays, British university students’ essays, and American university students’ essays: http s : / /www . [sent-158, score-0.295]

50 Similarly, we implemented the vector-based method with trigrams using the same frequency cutoff (but without smoothing). [sent-167, score-0.092]

51 The tree at the top is the Indo-European family tree drawn based on the figure shown in Crystal (1997). [sent-170, score-0.403]

52 The second and third trees are the cluster trees generated by the language model-based and vector-based methods, respectively. [sent-172, score-0.1]

53 The two languages belong to the North Germanic branch of the Germanic branch and thus are closely related. [sent-179, score-0.099]

54 ference between its cluster tree and the IndoEuropean family tree is that there are some mismatches within the Germanic and Slavic branches. [sent-185, score-0.439]

55 From these results, we can say that mother tongue interference is transferred into the 11 Englishes, strongly enough for reconstructing its language family tree, which we propose calling the interlanguage Indo-European family tree in English. [sent-188, score-1.645]

56 2 shows the experimental results with native and Japanese Englishes. [sent-190, score-0.195]

57 It shows that the same interlanguage Indo-European family tree was reconstructed as before. [sent-191, score-0.444]

58 More interestingly, native English was detached from the interlanguage Indo-European family tree contrary to the expectation that it would be attached to the Germanic branch because English is of course a member of the Germanic branch. [sent-192, score-0.61]

59 Interlagu eIndo-Eurpeanfm12ilytreEN1nag3tilsvheJEanpOgatlnhsie r31efJaEmnpgaislnyehs2 Figure 2: Experimental results with native and Japanese Englishes. [sent-194, score-0.195]

60 otherwise, native English would be included in the German branch. [sent-195, score-0.195]

61 Based on these results, we can further hypothesize as follows: interfamily distance > non-nativeness > intrafamily distance. [sent-199, score-0.14]

62 5 Discussion To get a better understanding of the interlanguage Indo-European family tree, we further explore lin- guistic features that explain well the above phenomena. [sent-200, score-0.298]

63 It is almost impossible to find someone who has a good knowledge of the 11 languages and their mother language interference in English writing. [sent-202, score-0.539]

64 Also, because we had access to a native speaker of Russian who had a good knowledge of English, we included Russian-English in our focus. [sent-209, score-0.195]

65 Second, we used a method for extracting interesting trigrams from the corpus data. [sent-211, score-0.092]

66 If we remove instances of a trigram from each set, the clustering tree involving 1141 the three may change. [sent-213, score-0.185]

67 For example, the removal of but the hers may result in a cluster tree merging French- and Russian-Englishes before Frenchand Spanish-Englishes. [sent-214, score-0.122]

68 We analyzed what trigrams had contributed to the clustering results with this approach. [sent-216, score-0.147]

69 Thus, the greater s is, the higher the chance that the cluster tree changes. [sent-221, score-0.122]

70 Therefore, we can obtain a list of interesting trigrams by sorting them according to s. [sent-222, score-0.092]

71 Table 2 shows the top 15 interesting trigrams where Di, Dj, and Dk are French-, Spanish-, and Russian-Englishes, respectively. [sent-226, score-0.092]

72 The list reveals that many of the trigrams contain the article a or the. [sent-228, score-0.18]

73 An exception is that oftrigrams containing the definite article in Bulgarian-English; it tends to be higher in Bulgarian-English than in the other Slavic Englishes. [sent-236, score-0.143]

74 Surprisingly and interestingly, however, it reflects the fact that Bulgarian does have the definite article but not the indefinite article (e. [sent-237, score-0.266]

75 Table 3 shows that the differences in article use exist even between the Italic and Germanic branches despite the fact that both have the indefinite and definite articles. [sent-246, score-0.226]

76 The definite article is used more frequently in the Italic-Englishes than in the Germanic Englishes (except for Dutch-English). [sent-252, score-0.143]

77 We speculate that this is perhaps because the Italic languages have a wider usage of the definite article such as its modification of possessive pronouns and proper nouns. [sent-253, score-0.18]

78 This corresponds to the fact that noun-noun compounds are less common in the Italic languages than in English and that instead, the of-phrase (NN of NN) is preferred (Swan and Smith, 2001). [sent-324, score-0.116]

79 instance, orange juice is expressed as juice of orange in the Italic languages (e. [sent-392, score-0.223]

80 In other words, the length tends to be shorter than in the others where we define the length as the number of consecutive repetitions of common nouns (for example, the length of orange juice is one because a noun is consecutively repeated once). [sent-401, score-0.126]

81 This tendency in the length of nounnoun compounds provides us with a crucial insight for native language identification, which we will Relative frequency of indefinite article (%) Figure 3: Distribution of articles. [sent-406, score-0.397]

82 (2005) work on native language identification and show that machine learning-based methods are effective. [sent-427, score-0.195]

83 The experimental results show that n-grams containing articles are predictive for identifying native languages. [sent-430, score-0.228]

84 This indicates that they should be used in the native language identification task. [sent-431, score-0.195]

85 Importantly, all n-grams containing articles should be used in the classifier unlike the previous methods that are based only on ngrams containing article errors. [sent-432, score-0.121]

86 ” In addition, the length of noun-noun compounds and the position of adverbs should also be considered in native language identification. [sent-435, score-0.306]

87 7 shows that the observed values in the French- English data very closely fit the theoretical proba11For comparison, we conducted a pilot study where we reconstructed a language family tree from English texts in European Parliament Proceedings Parallel Corpus (Europarl) (Koehn, 2011). [sent-442, score-0.435]

88 It turned out that the reconstructed tree was different from the canonical tree (available at http : / /web . [sent-443, score-0.232]

89 Consequently, Equation (5) should be useful in native language identi- fication. [sent-453, score-0.195]

90 In the domain of historical linguistics, researchers have used computational and corpusbased methods for reconstructing language family trees. [sent-455, score-0.319]

91 , 2005) apply clustering techniques to the task of language family tree reconstruction. [sent-459, score-0.372]

92 These methods reconstruct language family trees based on linguistic features that exist within words including lexical, phonological, and morphological features. [sent-461, score-0.33]

93 The experimental results in this paper suggest the possibility of the use of non-native texts for reconstructing language family trees. [sent-462, score-0.377]

94 7 Conclusions In this paper, we have shown that mother tongue interference is so strong that the relations between members of the Indo-European language family are preserved in English texts written by Indo-European language speakers. [sent-470, score-1.171]

95 To show this, we have used clustering to reconstruct a language family tree from 11 sets of non-native English texts. [sent-471, score-0.439]

96 It turned out that the reconstructed tree correctly groups them into the Italic, Germanic, and Slavic branches of the IndoEuropean family tree. [sent-472, score-0.425]

97 Based on the resulting trees, we have then hypothesized that the following relation holds in mother tongue interference: interfamily distance > non-nativeness > intrafamily distance. [sent-473, score-0.762]

98 Determining an author’s native language by mining a text for errors. [sent-562, score-0.195]

99 From bag of languages to family trees from noisy corpus. [sent-582, score-0.3]

100 A discriminant analysis of non-native speakers and native speakers of English. [sent-595, score-0.295]


similar papers computed by tfidf model

tfidf for this paper:

wordName wordTfidf (topN-words)

[('englishes', 0.349), ('tongue', 0.322), ('engilsh', 0.32), ('mother', 0.3), ('family', 0.231), ('interference', 0.202), ('native', 0.195), ('nn', 0.191), ('italic', 0.177), ('kita', 0.132), ('germanic', 0.126), ('slavic', 0.114), ('di', 0.108), ('essays', 0.1), ('trigrams', 0.092), ('reconstructing', 0.088), ('article', 0.088), ('tree', 0.086), ('transferred', 0.086), ('mi', 0.084), ('batagelj', 0.082), ('jj', 0.081), ('compounds', 0.079), ('eos', 0.076), ('mj', 0.073), ('interlanguage', 0.067), ('dras', 0.067), ('reconstruct', 0.067), ('reconstructed', 0.06), ('texts', 0.058), ('preserved', 0.058), ('ramat', 0.058), ('bos', 0.056), ('wong', 0.056), ('clustering', 0.055), ('nns', 0.055), ('definite', 0.055), ('rti', 0.054), ('subcorpus', 0.054), ('rb', 0.054), ('nonnative', 0.051), ('granger', 0.051), ('koppel', 0.051), ('speakers', 0.05), ('barbanc', 0.049), ('belly', 0.049), ('enright', 0.049), ('interfamily', 0.049), ('intrafamily', 0.049), ('nakhleh', 0.049), ('swan', 0.049), ('dj', 0.049), ('orange', 0.049), ('branches', 0.048), ('pr', 0.047), ('dk', 0.045), ('juice', 0.044), ('tongues', 0.044), ('trigram', 0.044), ('english', 0.043), ('distance', 0.042), ('japanese', 0.041), ('crftagger', 0.04), ('icle', 0.04), ('inflectional', 0.039), ('french', 0.039), ('phylogenetic', 0.038), ('languages', 0.037), ('cluster', 0.036), ('indoeuropean', 0.036), ('swedish', 0.035), ('europarl', 0.035), ('indefinite', 0.035), ('articles', 0.033), ('russian', 0.033), ('repetitions', 0.033), ('aarts', 0.033), ('altenberg', 0.033), ('elleg', 0.033), ('ivbp', 0.033), ('kroeber', 0.033), ('locness', 0.033), ('luay', 0.033), ('overuse', 0.033), ('ringe', 0.033), ('rtj', 0.033), ('rtk', 0.033), ('ryo', 0.033), ('strigram', 0.033), ('sugiura', 0.033), ('tandy', 0.033), ('tapper', 0.033), ('trtirtjrtk', 0.033), ('warnow', 0.033), ('adverbs', 0.032), ('strongly', 0.032), ('trees', 0.032), ('distances', 0.031), ('branch', 0.031), ('agglomerative', 0.031)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 1.0000011 299 acl-2013-Reconstructing an Indo-European Family Tree from Non-native English Texts

Author: Ryo Nagata ; Edward Whittaker

Abstract: Mother tongue interference is the phenomenon where linguistic systems of a mother tongue are transferred to another language. Although there has been plenty of work on mother tongue interference, very little is known about how strongly it is transferred to another language and about what relation there is across mother tongues. To address these questions, this paper explores and visualizes mother tongue interference preserved in English texts written by Indo-European language speakers. This paper further explores linguistic features that explain why certain relations are preserved in English writing, and which contribute to related tasks such as native language identification.

2 0.11817048 144 acl-2013-Explicit and Implicit Syntactic Features for Text Classification

Author: Matt Post ; Shane Bergsma

Abstract: Syntactic features are useful for many text classification tasks. Among these, tree kernels (Collins and Duffy, 2001) have been perhaps the most robust and effective syntactic tool, appealing for their empirical success, but also because they do not require an answer to the difficult question of which tree features to use for a given task. We compare tree kernels to different explicit sets of tree features on five diverse tasks, and find that explicit features often perform as well as tree kernels on accuracy and always in orders of magnitude less time, and with smaller models. Since explicit features are easy to generate and use (with publicly avail- able tools) , we suggest they should always be included as baseline comparisons in tree kernel method evaluations.

3 0.085495397 389 acl-2013-Word Association Profiles and their Use for Automated Scoring of Essays

Author: Beata Beigman Klebanov ; Michael Flor

Abstract: We describe a new representation of the content vocabulary of a text we call word association profile that captures the proportions of highly associated, mildly associated, unassociated, and dis-associated pairs of words that co-exist in the given text. We illustrate the shape of the distirbution and observe variation with genre and target audience. We present a study of the relationship between quality of writing and word association profiles. For a set of essays written by college graduates on a number of general topics, we show that the higher scoring essays tend to have higher percentages of both highly associated and dis-associated pairs, and lower percentages of mildly associated pairs of words. Finally, we use word association profiles to improve a system for automated scoring of essays.

4 0.067552805 98 acl-2013-Cross-lingual Transfer of Semantic Role Labeling Models

Author: Mikhail Kozhevnikov ; Ivan Titov

Abstract: Semantic Role Labeling (SRL) has become one of the standard tasks of natural language processing and proven useful as a source of information for a number of other applications. We address the problem of transferring an SRL model from one language to another using a shared feature representation. This approach is then evaluated on three language pairs, demonstrating competitive performance as compared to a state-of-the-art unsupervised SRL system and a cross-lingual annotation projection baseline. We also consider the contribution of different aspects of the feature representation to the performance of the model and discuss practical applicability of this method. 1 Background and Motivation Semantic role labeling has proven useful in many natural language processing tasks, such as question answering (Shen and Lapata, 2007; Kaisser and Webber, 2007), textual entailment (Sammons et al., 2009), machine translation (Wu and Fung, 2009; Liu and Gildea, 2010; Gao and Vogel, 2011) and dialogue systems (Basili et al., 2009; van der Plas et al., 2009). Multiple models have been designed to automatically predict semantic roles, and a considerable amount of data has been annotated to train these models, if only for a few more popular languages. As the annotation is costly, one would like to leverage existing resources to minimize the human effort required to construct a model for a new language. A number of approaches to the construction of semantic role labeling models for new languages have been proposed. On one end of the scale is unsupervised SRL, such as Grenager and Manning (2006), which requires some expert knowledge, but no labeled data. It clusters together arguments that should bear the same semantic role, but does not assign a particular role to each cluster. On the other end is annotating a new dataset from scratch. There are also intermediate options, which often make use of similarities between languages. This way, if an accurate model exists for one language, it should help simplify the construction of a model for another, related language. The approaches in this third group often use parallel data to bridge the gap between languages. Cross-lingual annotation projection systems (Pad o´ and Lapata, 2009), for example, propagate information directly via word alignment links. However, they are very sensitive to the quality of parallel data, as well as the accuracy of a sourcelanguage model on it. An alternative approach, known as cross-lingual model transfer, or cross-lingual model adaptation, consists of modifying a source-language model to make it directly applicable to a new language. This usually involves constructing a shared feature representation across the two languages. McDonald et al. (201 1) successfully apply this idea to the transfer of dependency parsers, using part-of- speech tags as the shared representation of words. A later extension of T ¨ackstr o¨m et al. (2012) enriches this representation with cross-lingual word clusters, considerably improving the performance. In the case of SRL, a shared representation that is purely syntactic is likely to be insufficient, since structures with different semantics may be realized by the same syntactic construct, for example “in August” vs “in Britain”. However with the help of recently introduced cross-lingual word represen1190 Proce dingsS o f ita h,e B 5u1lgsta Arinan,u Aaulg Musete 4ti-n9g 2 o0f1 t3h.e ? Ac s2s0o1ci3a Atiosnso fcoirat Cio nm foprut Caotimonpaulta Lti nognuails Lti cnsg,u piasgteics 1 90–120 , tations, such as the cross-lingual clustering mentioned above or cross-lingual distributed word representations of Klementiev et al. (2012), we may be able to transfer models of shallow semantics in a similar fashion. In this work we construct a shared feature representation for a pair of languages, employing crosslingual representations of syntactic and lexical information, train a semantic role labeling model on one language and apply it to the other one. This approach yields an SRL model for a new language at a very low cost, effectively requiring only a source language model and parallel data. We evaluate on five (directed) language pairs EN-ZH, ZH-EN, EN-CZ, CZ-EN and EN-FR, where EN, FR, CZ and ZH denote English, French, Czech and Chinese, respectively. The transferred model is compared against two baselines: an unsupervised SRL system and a model trained on the output of a cross-lingual annotation projection system. In the next section we will describe our setup, then in section 3 present the shared feature representation we use, discuss the evaluation data and other technical aspects in section 4, present the results and conclude with an overview of related work. – 2 Setup The purpose of the study is not to develop a yet another semantic role labeling system any existing SRL system can (after some modification) be used in this setup but to assess the practical applicability of cross-lingual model transfer to this – – problem, compare it against the alternatives and identify its strong/weak points depending on a particular setup. 2.1 Semantic Role Labeling Model We consider the dependency-based version of semantic role labeling as described in Haji cˇ et al. (2009) and transfer an SRL model from one language to another. We only consider verbal predicates and ignore the predicate disambiguation stage. We also assume that the predicate identification information is available in most languages it can be obtained using a relatively simple heuristic based on part-of-speech tags. The model performs argument identification and classification (Johansson and Nugues, 2008) separately in a pipeline first each candidate is classified as being or not being a head of an argument phrase with respect to the predicate in question and then each of the arguments is assigned a role from a given inventory. The model is factorized over arguments the decisions regarding the classification of different arguments are made in– – – dependently of each other. With respect to the use of syntactic annotation we consider two options: using an existing dependency parser for the target language and obtaining one by means of cross-lingual transfer (see section 4.2). Following McDonald et al. (201 1), we assume that a part-of-speech tagger is available for the target language. 2.2 SRL in the Low-resource Setting Several approaches have been proposed to obtain an SRL model for a new language with little or no manual annotation. Unsupervised SRL models (Lang and Lapata, 2010) cluster the arguments of predicates in a given corpus according to their semantic roles. The performance of such models can be impressive, especially for those languages where semantic roles correlate strongly with syntactic relation of the argument to its predicate. However, assigning meaningful role labels to the resulting clusters requires additional effort and the model’s parameters generally need some adjustment for every language. If the necessary resources are already available for a closely related language, they can be utilized to facilitate the construction of a model for the target language. This can be achieved either by means of cross-lingual annotation projection (Yarowsky et al., 2001) or by cross-lingual model transfer (Zeman and Resnik, 2008). This last approach is the one we are considering in this work, and the other two options are treated as baselines. The unsupervised model will be further referred to as UNSUP and the projection baseline as PROJ. 2.3 Evaluation Measures We use the F1 measure as a metric for the argument identification stage and accuracy as an aggregate measure of argument classification performance. When comparing to the unsupervised SRL system the clustering evaluation measures are used instead. These are purity and collocation 1191 N1Ximajx|Gj∩ Ci| CO =N1Xjmiax|Gj∩ Ci|, PU = where Ci is the set of arguments in the i-th induced cluster, Gj is the set of arguments in the jth gold cluster and N is the total number of arguments. We report the harmonic mean ofthe two (Lang and Lapata, 2011) and denote it F1c to avoid confusing it with the supervised metric. 3 Model Transfer The idea of this work is to abstract the model away from the particular source language and apply it to a new one. This setup requires that we use the same feature representation for both languages, for example part-of-speech tags and dependency relation labels should be from the same inventory. Some features are not applicable to certain lan- guages because the corresponding phenomena are absent in them. For example, consider a strongly inflected language and an analytic one. While the latter can usually convey the information encoded in the word form in the former one (number, gender, etc.), finding a shared feature representation for such information is non-trivial. In this study we will confine ourselves to those features that are applicable to all languages in question, namely: part-of-speech tags, syntactic dependency structures and representations of the word’s identity. 3.1 Lexical Information We train a model on one language and apply it to a different one. In order for this to work, the words of the two languages have to be mapped into a common feature space. It is also desirable that closely related words from both languages have similar representations in this space. Word mapping. The first option is simply to use the source language words as the shared representation. Here every source language word would have itself as its representation and every target word would map into a source word that corresponds to it. In other words, we supply the model with a gloss of the target sentence. The mapping (bilingual dictionary) we use is derived from a word-aligned parallel corpus, by identifying, for each word in the target language, the word in the source language it is most often aligned to. Cross-lingual clusters. There is no guarantee that each of the words in the evaluation data is present in our dictionary, nor that the corresponding source-language word is present in the training data, so the model would benefit from the ability to generalize over closely related words. This can, for example, be achieved by using cross-lingual word clusters induced in T ¨ackstr o¨m et al. (2012). We incorporate these clusters as features into our model. 3.2 Syntactic Information Part-of-speech Tags. We map part-of-speech tags into the universal tagset following Petrov et al. (2012). This may have a negative effect on the performance of a monolingual model, since most part-of-speech tagsets are more fine-grained than the universal POS tags considered here. For example Penn Treebank inventory contains 36 tags and the universal POS tagset only 12. Since the finergrained POS tags often reflect more languagespecific phenomena, however, they would only be useful for very closely related languages in the cross-lingual setting. The universal part-of-speech tags used in evaluation are derived from gold-standard annotation for all languages except French, where predicted ones had to be used instead. Dependency Structure. Another important aspect of syntactic information is the dependency structure. Most dependency relation inventories are language-specific, and finding a shared representation for them is a challenging problem. One could map dependency relations into a simplified form that would be shared between languages, as it is done for part-of-speech tags in Petrov et al. (2012). The extent to which this would be useful, however, depends on the similarity of syntactic-semantic in– terfaces of the languages in question. In this work we discard the dependency relation labels where the inventories do not match and only consider the unlabeled syntactic dependency graph. Some discrepancies, such as variations in attachment order, may be present even there, but this does not appear to be the case with the datasets we use for evaluation. If a target language is poor in resources, one can obtain a dependency parser for the target language by means of cross-lingual model transfer (Zeman and Resnik, 2008). We 1192 take this into account and evaluate both using the original dependency structures and the ones obtained by means of cross-lingual model transfer. 3.3 The Model The model we use is based on that of Bj ¨orkelund et al. (2009). It is comprised of a set of linear classifiers trained using Liblinear (Fan et al., 2008). The feature model was modified to accommodate the cross-lingual cluster features and the reranker component was not used. We do not model the interaction between different argument roles in the same predicate. While this has been found useful, in the cross-lingual setup one has to be careful with the assumptions made. For example, modeling the sequence of roles using a Markov chain (Thompson et al., 2003) may not work well in the present setting, especially between distant languages, as the order or arguments is not necessarily preserved. Most constraints that prove useful for SRL (Chang et al., 2007) also require customization when applied to a new language, and some rely on languagespecific resources, such as a valency lexicon. Taking into account the interaction between different arguments of a predicate is likely to improve the performance of the transferred model, but this is outside the scope of this work. 3.4 Feature Selection Compatibility of feature representations is necessary but not sufficient for successful model transfer. We have to make sure that the features we use are predictive of similar outcomes in the two languages as well. Depending on the pair of languages in question, different aspects of the feature representation will retain or lose their predictive power. We can be reasonably certain that the identity of an argument word is predictive of its semantic role in any language, but it might or might not be true of, for example, the word directly preceding the argument word. It is therefore important to pre- SCPDGylOespoSntreslTabunc1lra:obsFel-daitnguplrdoaeusntpagd-elronwfu-dcsopeyrnsd c.eylafguhtorsia mepgnrhs vent the model from capturing overly specific aspects of the source language, which we do by confining the model to first-order features. We also avoid feature selection, which, performed on the source language, is unlikely to help the model to better generalize to the target one. The experiments confirm that feature selection and the use of second-order features degrade the performance of the transferred model. 3.5 Feature Groups For each word, we use its part-of-speech tag, cross-lingual cluster id, word identity (glossed, when evaluating on the target language) and its dependency relation to its parent. Features associated with an argument word include the attributes of the predicate word, the argument word, its parent, siblings and children, and the words directly preceding and following it. Also included are the sequences of part-of-speech tags and dependency relations on the path between the predicate and the argument. Since we are also interested in the impact of different aspects of the feature representation, we divide the features into groups as summarized in table 1 and evaluate their respective contributions to the performance of the model. If a feature group is enabled the model has access to the corre– sponding source of information. For example, if only POS group is enabled, the model relies on the part-of-speech tags of the argument, the predicate and the words to the right and left of the argument word. If Synt is enabled too, it also uses the POS tags of the argument’s parent, children and siblings. Word order information constitutes an implicit group that is always available. It includes the Pos it ion feature, which indicates whether the argument is located to the left or to the right of the predicate, and allows the model to look up the attributes of the words directly preceding and following the argument word. The model we compare against the baselines uses all applicable feature groups (Deprel is only used in EN-CZ and CZ-EN experiments with original syntax). 4 Evaluation 4.1 Datasets and Preprocessing Evaluation of the cross-lingual model transfer requires a rather specific kind of dataset. Namely, the data in both languages has to be annotated 1193 with the same set of semantic roles following the same (or compatible) guidelines, which is seldom the case. We have identified three language pairs for which such resources are available: EnglishChinese, English-Czech and English-French. The evaluation datasets for English and Chinese are those from the CoNLL Shared Task 2009 (Haji ˇc et al., 2009) (henceforth CoNLL-ST). Their annotation in the CoNLL-ST is not identical, but the guidelines for “core” semantic roles are similar (Kingsbury et al., 2004), so we evaluate only on core roles here. The data for the second language pair is drawn from the Prague Czech-English Dependency Treebank 2.0 (Haji ˇc et al., 2012), which we converted to a format similar to that of CoNLL-ST1 . The original annotation uses the tectogrammatical representation (Haji ˇc, 2002) and an inventory of semantic roles (or functors), most of which are interpretable across various predicates. Also note that the syntactic anno- tation of English and Czech in PCEDT 2.0 is quite similar (to the extent permitted by the difference in the structure of the two languages) and we can use the dependency relations in our experiments. For English-French, the English CoNLL-ST dataset was used as a source and the model was evaluated on the manually annotated dataset from van der Plas et al. (201 1). The latter contains one thousand sentences from the French part ofthe Europarl (Koehn, 2005) corpus, annotated with semantic roles following an adapted version of PropBank (Palmer et al., 2005) guidelines. The authors perform annotation projection from English to French, using a joint model of syntax and semantics and employing heuristics for filtering. We use a model trained on the output of this projection system as one of the baselines. The evaluation dataset is relatively small in this case, so we perform the transfer only one-way, from English to French. The part-of-speech tags in all datasets were replaced with the universal POS tags of Petrov et al. (2012). For Czech, we have augmented the map- pings to account for the tags that were not present in the datasets from which the original mappings were derived. Namely, tag “t” is mapped to “VERB” and “Y” to “PRON”. We use parallel data to construct a bilingual dictionary used in word mapping, as well as in the projection baseline. For English-Czech – 1see http://www.ml4nlp.de/code-and-data/treex2conll and English-French, the data is drawn from Europarl (Koehn, 2005), for English-Chinese from MultiUN (Eisele and Chen, 2010). The word alignments were obtained using GIZA++ (Och and Ney, 2003) and the intersection heuristic. – 4.2 Syntactic Transfer In the low-resource setting, we cannot always rely on the availability of an accurate dependency parser for the target language. If one is not available, the natural solution would be to use crosslingual model transfer to obtain it. Unfortunately, the models presented in the previous work, such as Zeman and Resnik (2008), McDonald et al. (201 1) and T ¨ackstr o¨m et al. (2012), were not made available, so we reproduced the direct transfer algorithm of McDonald et al. (201 1), using Malt parser (Nivre, 2008) and the same set of features. We did not reimplement the projected transfer algorithm, however, and used the default training procedure instead of perceptron-based learning. The dependency structure thus obtained is, of course, only a rough approximation even a much more sophisticated algorithm may not perform well when transferring syntax between such languages as Czech and English, given the inherent difference in their structure. The scores are shown in table 2. We will henceforth refer to the syntactic annotations that were provided with the datasets as original, as opposed to the annotations obtained by means of syntactic transfer. – 4.3 Baselines Unsupervised Baseline: We are using a version of the unsupervised semantic role induction system of Titov and Klementiev (2012a) adapted to SetupUAS, % Table2:SyntaciE C ZcN HNt- rE ZaCFnN HZRsfer34 692567acuracy,unlabe dat- tachment score (percent). Note that in case of French we evaluate against the output of a supervised system, since manual annotation is not available for this dataset. This score does not reflect the true performance of syntactic transfer. 1194 the shared feature representation considered in order to make the scores comparable with those of the transfer model and, more importantly, to enable evaluation on transferred syntax. Note that the original system, tailored to a more expressive language-specific syntactic representation and equipped with heuristics to identify active/passive voice and other phenomena, achieves higher scores than those we report here. Projection Baseline: The projection baseline we use for English-Czech and English-Chinese is a straightforward one: we label the source side of a parallel corpus using the source-language model, then identify those verbs on the target side that are aligned to a predicate, mark them as predicates and propagate the argument roles in the same fashion. A model is then trained on the resulting training data and applied to the test set. For English-French we instead use the output of a fully featured projection model of van der Plas et al. (201 1), published in the CLASSiC project. 5 Results In order to ensure that the results are consistent, the test sets, except for the French one, were partitioned into five equal parts (of 5 to 10 thousand sentences each, depending on the dataset) and the evaluation performed separately on each one. All evaluation figures for English, Czech or Chinese below are the average values over the five subsets. In case of French, the evaluation dataset is too small to split it further, so instead we ran the evaluation five times on a randomly selected 80% sample of the evaluation data and averaged over those. In both cases the results are consistent over the subsets, the standard deviation does not exceed 0.5% for the transfer system and projection baseline and 1% for the unsupervised system. 5.1 Argument Identification We summarize the results in table 3. Argument identification is known to rely heavily on syntactic information, so it is unsurprising that it proves inaccurate when transferred syntax is used. Our simple projection baseline suffers from the same problem. Even with original syntactic information available, the performance of argument identification is moderate. Note that the model of (van der Plas et al., 2011), though relying on more expressive syntax, only outperforms the transferred system by 3% (F1) on this task. SetupSyntaxTRANSPROJ ZEC NH Z- EFCZNRHt r a n s 3462 1. 536 142 35. 4269 Table3EZ C:N H- CFEZANHZRrgumeon rt ig identf56 7ic13 a. t27903ion,21569t10ra. 3976nsferd model vs. projection baseline, F1. Most unsupervised SRL approaches assume that the argument identification is performed by some external means, for example heuristically (Lang and Lapata, 2011). Such heuristics or unsupervised approaches to argument identification (Abend et al., 2009) can also be used in the present setup. 5.2 Argument Classification In the following tables, TRANS column contains the results for the transferred system, UNSUP for the unsupervised baseline and PROJ for projection baseline. We highlight in bold the higher score where the difference exceeds twice the maximum of the standard deviation estimates of the two results. Table 4 presents the unsupervised evaluation results. Note that the unsupervised model performs as well as the transferred one or better where the – – SetupSyntaxTRANSUNSUP ZEC NH Z- EFCZNRHt r a n s 768 93648. 34627 6 5873. 1769 TableEZ C4NHZ:- FCEZANHZRrgumoe nr itg clasi78 fi94 3c. a25136tion,8 7 r9a4263n. 07 sferd model vs. unsupervised baseline in terms of the clustering metric F1c (see section 2.3). 1195 SetupSyntaxTRANSPROJ ZEC NH Z- EFCZNRHt r a n s 657 053. 1 36456419. 372 Table5EZ C:N H- CFEZANHZRrgumeon rt ig clasif657ic1936a. t170 ion,65 9t3804ra. 20847nsferd model vs. projection baseline, accuracy. original syntactic dependencies are available. In the more realistic scenario with transferred syn- tax, however, the transferred model proves more accurate. In table 5 we compare the transferred system with the projection baseline. It is easy to see that the scores vary strongly depending on the language pair, due to both the difference in the annotation scheme used and the degree of relatedness between the languages. The drop in performance when transferring the model to another language is large in every case, though, see table 6. SetupTargetSource Table6:MoCEZdHeNZ l- FECaZNRcH urac67 y53169o. 017nthes87 o25670u. r1245ceandtrge language using original syntax. The source language scores for English vary between language pairs because of the difference in syntactic annotation and role subset used. We also include the individual F1 scores for the top-10 most frequent labels for EN-CZ transfer with original syntax in table 7. The model provides meaningful predictions here, despite low overall accuracy. Most of the labels2 are self-explanatory: Patient (PAT), Actor (ACT), Time (TWHEN), Effect (EFF), Location (LOC), Manner (MANN), Addressee (ADDR), Extent (EXT). CPHR marks the 2http://ufal.mff.cuni.cz/∼toman/pcedt/en/functors.html LabelFreq.F1Re.Pr. recall and precision for the top-10 most frequent roles. nominal part of a complex predicate, as in “to have [a plan]CPHR”, and DIR3 indicates destination. 5.3 Additional Experiments We now evaluate the contribution of different aspects of the feature representation to the performance of the model. Table 8 contains the results for English-French. FeaturesOrigTrans ferent feature subsets, using original and transferred syntactic information. The fact that the model performs slightly better with transferred syntax may be explained by two factors. Firstly, as we already mentioned, the original syntactic annotation is also produced automatically. Secondly, in the model transfer setup it is more important how closely the syntacticsemantic interface on the target side resembles that on the source side than how well it matches the “true” structure of the target language, and in this respect a transferred dependency parser may have an advantage over one trained on target-language data. The high impact of the Glos s features here 1196 may be partly attributed to the fact that the mapping is derived from the same corpus as the evaluation data Europarl (Koehn, 2005) and partly by the similarity between English and French in terms of word order, usage of articles and prepositions. The moderate contribution of the crosslingual cluster features are likely due to the insufficient granularity of the clustering for this task. For more distant language pairs, the contributions of individual feature groups are less interpretable, so we only highlight a few observations. First of all, both EN-CZ and CZ-EN benefit noticeably from the use of the original syntactic annotation, including dependency relations, but not from the transferred syntax, most likely due to the low syntactic transfer performance. Both perform better when lexical information is available, although – – the improvement is not as significant as in the case of French only up to 5%. The situation with Chinese is somewhat complicated in that adding lexical information here fails to yield an improvement in terms of the metric considered. This is likely due to the fact that we consider only the core roles, which can usually be predicted with high accuracy based on syntactic information alone. – 6 Related Work Development of robust statistical models for core NLP tasks is a challenging problem, and adaptation of existing models to new languages presents a viable alternative to exhaustive annotation for each language. Although the models thus obtained are generally imperfect, they can be further refined for a particular language and domain using techniques such as active learning (Settles, 2010; Chen et al., 2011). Cross-lingual annotation projection (Yarowsky et al., 2001) approaches have been applied ex- tensively to a variety of tasks, including POS tagging (Xi and Hwa, 2005; Das and Petrov, 2011), morphology segmentation (Snyder and Barzilay, 2008), verb classification (Merlo et al., 2002), mention detection (Zitouni and Florian, 2008), LFG parsing (Wr o´blewska and Frank, 2009), information extraction (Kim et al., 2010), SRL (Pad o´ and Lapata, 2009; van der Plas et al., 2011; Annesi and Basili, 2010; Tonelli and Pianta, 2008), dependency parsing (Naseem et al., 2012; Ganchev et al., 2009; Smith and Eisner, 2009; Hwa et al., 2005) or temporal relation prediction (Spreyer and Frank, 2008). Interestingly, it has also been used to propagate morphosyntactic information between old and modern versions of the same language (Meyer, 2011). Cross-lingual model transfer methods (McDonald et al., 2011; Zeman and Resnik, 2008; Durrett et al., 2012; Søgaard, 2011; Lopez et al., 2008) have also been receiving much attention recently. The basic idea behind model transfer is similar to that of cross-lingual annotation projection, as we can see from the way parallel data is used in, for example, McDonald et al. (201 1). A crucial component of direct transfer approaches is the unified feature representation. There are at least two such representations of lexical information (Klementiev et al., 2012; T ¨ackstr o¨m et al., 2012), but both work on word level. This makes it hard to account for phenomena that are expressed differently in the languages considered, for example the syntactic function of a certain word may be indicated by a preposition, inflection or word order, depending on the language. Accurate representation of such information would require an extra level of abstraction (Haji ˇc, 2002). A side-effect ofusing adaptation methods is that we are forced to use the same annotation scheme for the task in question (SRL, in our case), which in turn simplifies the development of cross-lingual tools for downstream tasks. Such representations are also likely to be useful in machine translation. Unsupervised semantic role labeling methods (Lang and Lapata, 2010; Lang and Lapata, 2011; Titov and Klementiev, 2012a; Lorenzo and Cerisara, 2012) also constitute an alternative to cross-lingual model transfer. For an overview of of semi-supervised approaches we refer the reader to Titov and Klementiev (2012b). 7 Conclusion We have considered the cross-lingual model transfer approach as applied to the task of semantic role labeling and observed that for closely related languages it performs comparably to annotation projection approaches. It allows one to quickly construct an SRL model for a new language without manual annotation or language-specific heuristics, provided an accurate model is available for one of the related languages along with a certain amount of parallel data for the two languages. While an1197 notation projection approaches require sentenceand word-aligned parallel data and crucially depend on the accuracy of the syntactic parsing and SRL on the source side of the parallel corpus, cross-lingual model transfer can be performed using only a bilingual dictionary. Unsupervised SRL approaches have their advantages, in particular when no annotated data is available for any of the related languages and there is a syntactic parser available for the target one, but the annotation they produce is not always sufficient. In applications such as Information Retrieval it is preferable to have precise labels, rather than just clusters of arguments, for example. Also note that when applying cross-lingual model transfer in practice, one can improve upon the performance of the simplistic model we use for evaluation, for example by picking the features manually, taking into account the properties of the target language. Domain adaptation techniques can also be employed to adjust the model to the target language. Acknowledgments The authors would like to thank Alexandre Klementiev and Ryan McDonald for useful suggestions and T ¨ackstr o¨m et al. (2012) for sharing the cross-lingual word representations. This research is supported by the MMCI Cluster of Excellence. References Omri Abend, Roi Reichart, and Ari Rappoport. 2009. Unsupervised argument identification for semantic role labeling. In Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP, ACL ’09, pages 28–36, Stroudsburg, PA, USA. Association for Computational Linguistics. Paolo Annesi and Roberto Basili. 2010. Cross-lingual alignment of FrameNet annotations through hidden Markov models. In Proceedings of the 11th international conference on Computational Linguistics and Intelligent Text Processing, CICLing’ 10, pages 12– 25, Berlin, Heidelberg. Springer-Verlag. Roberto Basili, Diego De Cao, Danilo Croce, Bonaventura Coppola, and Alessandro Moschitti. 2009. Cross-language frame semantics transfer in bilingual corpora. In Alexander F. Gelbukh, editor, Proceedings of the 10th International Conference on Computational Linguistics and Intelligent Text Pro- cessing, pages 332–345. Anders Bj ¨orkelund, Love Hafdell, and Pierre Nugues. 2009. Multilingual semantic role labeling. In Proceedings of the Thirteenth Conference on Computational Natural Language Learning (CoNLL 2009): Shared Task, pages 43–48, Boulder, Colorado, June. Association for Computational Linguistics. Ming-Wei Chang, Lev Ratinov, and Dan Roth. 2007. Guiding semi-supervision with constraint-driven learning. In ACL. Chenhua Chen, Alexis Palmer, and Caroline Sporleder. 2011. Enhancing active learning for semantic role labeling via compressed dependency trees. In Proceedings of 5th International Joint Conference on Natural Language Processing, pages 183–191, Chiang Mai, Thailand, November. Asian Federation of Natural Language Processing. Dipanjan Das and Slav Petrov. 2011. Unsupervised part-of-speech tagging with bilingual graph-based projections. Proceedings of the Association for Computational Linguistics. Greg Durrett, Adam Pauls, and Dan Klein. 2012. Syntactic transfer using a bilingual lexicon. In Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pages 1–1 1, Jeju Island, Korea, July. Association for Computational Linguistics. Andreas Eisele and Yu Chen. 2010. MultiUN: A multilingual corpus from United Nation documents. In Proceedings of the Seventh International Conference on Language Resources and Evaluation (LREC’10). European Language Resources Association (ELRA). Rong-En Fan, Kai-Wei Chang, Cho-Jui Hsieh, XiangRui Wang, and Chih-Jen Lin. 2008. LIBLINEAR: A library for large linear classification. Journal of Machine Learning Research, 9: 1871–1874. Kuzman Ganchev, Jennifer Gillenwater, and Ben Taskar. 2009. Dependency grammar induction via bitext projection constraints. In Proceedings of the 47th Annual Meeting of the ACL, pages 369–377, Stroudsburg, PA, USA. Association for Computational Linguistics. Qin Gao and Stephan Vogel. 2011. Corpus expansion for statistical machine translation with semantic role label substitution rules. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, pages 294–298, Portland, Oregon, USA. Trond Grenager and Christopher D. Manning. 2006. Unsupervised discovery of a statistical verb lexicon. In Proceedings of EMNLP. Jan Haji cˇ. 2002. Tectogrammatical representation: Towards a minimal transfer in machine translation. In Robert Frank, editor, Proceedings of the 6th International Workshop on Tree Adjoining Grammars 1198 and Related Frameworks (TAG+6), pages 216— 226, Venezia. Universita di Venezia. Jan Haji cˇ, Massimiliano Ciaramita, Richard Johansson, Daisuke Kawahara, Maria Ant o`nia Mart ı´, Llu ı´s M `arquez, Adam Meyers, Joakim Nivre, Sebastian Pad o´, Jan Sˇt eˇp a´nek, Pavel Stra nˇ a´k, Mihai Surdeanu, Nianwen Xue, and Yi Zhang. 2009. The CoNLL2009 shared task: Syntactic and semantic dependencies in multiple languages. In Proceedings of the Thirteenth Conference on Computational Natural Language Learning (CoNLL 2009): Shared Task, pages 1–18, Boulder, Colorado. Jan Haji cˇ, Eva Haji cˇov a´, Jarmila Panevov a´, Petr Sgall, Ond ˇrej Bojar, Silvie Cinkov´ a, Eva Fuˇ c ´ıkov a´, Marie Mikulov a´, Petr Pajas, Jan Popelka, Ji ˇr´ ı Semeck´ y, Jana Sˇindlerov a´, Jan Sˇt eˇp a´nek, Josef Toman, Zde nˇka Ure sˇov a´, and Zden eˇk Zˇabokrtsk y´. 2012. Announcing Prague Czech-English dependency treebank 2.0. In Nicoletta Calzolari (Conference Chair), Khalid Choukri, Thierry Declerck, Mehmet U gˇur Doˇ gan, Bente Maegaard, Joseph Mariani, Jan Odijk, and Stelios Piperidis, editors, Proceedings of the Eight International Conference on Language Resources and Evaluation (LREC’12), Istanbul, Turkey, May. European Language Resources Association (ELRA). Rebecca Hwa, Philip Resnik, Amy Weinberg, Clara Cabezas, and Okan Kolak. 2005. Bootstrapping parsers via syntactic projection across parallel text. Natural Language Engineering, 11(3):3 11–325. Richard Johansson and Pierre Nugues. 2008. Dependency-based semantic role labeling of PropBank. In Proceedings of the 2008 Conference on Empirical Methods in Natural Language Processing, pages 69–78, Honolulu, Hawaii. Michael Kaisser and Bonnie Webber. 2007. Question answering based on semantic roles. In ACL Workshop on Deep Linguistic Processing. Seokhwan Kim, Minwoo Jeong, Jonghoon Lee, and Gary Geunbae Lee. 2010. A cross-lingual annotation projection approach for relation detection. In Proceedings of the 23rd International Conference on Computational Linguistics, COLING ’ 10, pages 564–571, Stroudsburg, PA, USA. Association for Computational Linguistics. Paul Kingsbury, Nianwen Xue, and Martha Palmer. 2004. Propbanking in parallel. In In Proceedings of the Workshop on the Amazing Utility of Parallel and Comparable Corpora, in conjunction with LREC’04. Alexandre Klementiev, Ivan Titov, and Binod Bhattarai. 2012. Inducing crosslingual distributed representations of words. In Proceedings of the International Conference on Computational Linguistics (COLING), Bombay, India. Philipp Koehn. 2005. Europarl: A parallel corpus for statistical machine translation. In Conference Proceedings: the tenth Machine Translation Summit, pages 79–86, Phuket, Thailand. AAMT. Joel Lang and Mirella Lapata. 2010. Unsupervised induction of semantic roles. In Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, pages 939–947, Los Angeles, California, June. Association for Computational Linguistics. Joel Lang and Mirella Lapata. 2011. Unsupervised semantic role induction via split-merge clustering. In Proc. of Annual Meeting of the Association for Computational Linguistics (ACL). Ding Liu and Daniel Gildea. 2010. Semantic role features for machine translation. In Proceedings of the 23rd International Conference on Computational Linguistics (Coling 2010), Beijing, China. Adam Lopez, Daniel Zeman, Michael Nossal, Philip Resnik, and Rebecca Hwa. 2008. Cross-language parser adaptation between related languages. In IJCNLP-08 Workshop on NLP for Less Privileged Languages, pages 35–42, Hyderabad, India, January. Alejandra Lorenzo and Christophe Cerisara. 2012. Unsupervised frame based semantic role induction: application to French and English. In Proceedings of the ACL 2012 Joint Workshop on Statistical Parsing and Semantic Processing of Morphologically Rich Languages, pages 30–35, Jeju, Republic of Korea, July. Association for Computational Linguistics. Ryan McDonald, Slav Petrov, and Keith Hall. 2011. Multi-source transfer of delexicalized dependency parsers. In Proceedings of the Conference on Empirical Methods in Natural Language Processing, EMNLP ’ 11, pages 62–72, Stroudsburg, PA, USA. Association for Computational Linguistics. Paola Merlo, Suzanne Stevenson, Vivian Tsang, and Gianluca Allaria. 2002. A multi-lingual paradigm for automatic verb classification. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics (ACL’02), pages 207– 214, Philadelphia, PA. Roland Meyer. 2011. New wine in old wineskins?– Tagging old Russian via annotation projection from modern translations. Russian Linguistics, 35(2):267(15). Tahira Naseem, Regina Barzilay, and Amir Globerson. 2012. Selective sharing for multilingual dependency parsing. In Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics, pages 629–637, Jeju Island, Korea, July. Association for Computational Linguistics. Joakim Nivre. 2008. Algorithms for deterministic incremental dependency parsing. Comput. Linguist., 34(4):513–553, December. 1199 Franz Josef Och and Hermann Ney. 2003. A systematic comparison of various statistical alignment models. Computational Linguistics, 29(1). Sebastian Pad o´ and Mirella Lapata. 2009. Crosslingual annotation projection for semantic roles. Journal of Artificial Intelligence Research, 36:307– 340. Martha Palmer, Daniel Gildea, and Paul Kingsbury. 2005. The Proposition Bank: An annotated corpus of semantic roles. Computational Linguistics, 31:71–105. Slav Petrov, Dipanjan Das, and Ryan McDonald. 2012. A universal part-of-speech tagset. In Proceedings of LREC, May. Mark Sammons, Vinod Vydiswaran, Tim Vieira, Nikhil Johri, Ming wei Chang, Dan Goldwasser, Vivek Srikumar, Gourab Kundu, Yuancheng Tu, Kevin Small, Joshua Rule, Quang Do, and Dan Roth. 2009. Relation alignment for textual entailment recognition. In Text Analysis Conference (TAC). Burr Settles. 2010. Active learning literature survey. Computer Sciences Technical Report, 1648. Dan Shen and Mirella Lapata. 2007. Using semantic roles to improve question answering. In EMNLP. David A Smith and Jason Eisner. 2009. Parser adaptation and projection with quasi-synchronous grammar features. In Proceedings of the 2009 Confer- ence on Empirical Methods in Natural Language Processing, pages 822–831. Association for Computational Linguistics. Benjamin Snyder and Regina Barzilay. 2008. Crosslingual propagation for morphological analysis. In Proceedings of the 23rd national conference on Artificial intelligence. Anders Søgaard. 2011. Data point selection for crosslanguage adaptation of dependency parsers. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, volume 2 of HLT ’11, pages 682–686, Stroudsburg, PA, USA. Association for Computational Linguistics. Kathrin Spreyer and Anette Frank. 2008. Projectionbased acquisition of a temporal labeller. Proceedings of IJCNLP 2008. Oscar T¨ ackstr o¨m, Ryan McDonald, and Jakob Uszkoreit. 2012. Cross-lingual word clusters for direct transfer of linguistic structure. In Proc. of the Annual Meeting of the North American Association of Computational Linguistics (NAACL), pages 477– 487, Montr ´eal, Canada. Cynthia A. Thompson, Roger Levy, and Christopher D. Manning. 2003. A generative model for seman- tic role labeling. In Proceedings of the 14th European Conference on Machine Learning, ECML 2003, pages 397–408, Dubrovnik, Croatia. Ivan Titov and Alexandre Klementiev. 2012a. A Bayesian approach to unsupervised semantic role induction. In Proc. of European Chapter of the Association for Computational Linguistics (EACL). Ivan Titov and Alexandre Klementiev. 2012b. Semisupervised semantic role labeling: Approaching from an unsupervised perspective. In Proceedings of the International Conference on Computational Linguistics (COLING), Bombay, India, December. Sara Tonelli and Emanuele Pianta. 2008. Frame information transfer from English to Italian. In Proceedings of LREC 2008. Lonneke van der Plas, James Henderson, and Paola Merlo. 2009. Domain adaptation with artificial data for semantic parsing of speech. In Proc. 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, pages 125–128, Boulder, Colorado. Lonneke van der Plas, Paola Merlo, and James Henderson. 2011. Scaling up automatic cross-lingual semantic role annotation. In Proceedings of the 49th Annual Meeting of the Association for Computa- tional Linguistics: Human Language Technologies, HLT ’ 11, pages 299–304, Stroudsburg, PA, USA. Association for Computational Linguistics. Alina Wr o´blewska and Anette Frank. 2009. Crosslingual projection of LFG F-structures: Building an F-structure bank for Polish. In Eighth International Workshop on Treebanks and Linguistic Theories, page 209. Dekai Wu and Pascale Fung. 2009. Can semantic role labeling improve SMT? In Proceedings of 13th Annual Conference of the European Association for Machine Translation (EAMT 2009), Barcelona. Chenhai Xi and Rebecca Hwa. 2005. A backoff model for bootstrapping resources for non-English languages. In Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing, pages 85 1–858, Stroudsburg, PA, USA. David Yarowsky, Grace Ngai, and Ricahrd Wicentowski. 2001. Inducing multilingual text analysis tools via robust projection across aligned corpora. In Proceedings of Human Language Technology Conference. Daniel Zeman and Philip Resnik. 2008. Crosslanguage parser adaptation between related lan- guages. In Proceedings of the IJCNLP-08 Workshop on NLP for Less Privileged Languages, pages 35– 42, Hyderabad, India, January. Asian Federation of Natural Language Processing. Imed Zitouni and Radu Florian. 2008. Mention detection crossing the language barrier. In Proceedings of the Conference on Empirical Methods in Natural Language Processing. 1200

5 0.067338206 246 acl-2013-Modeling Thesis Clarity in Student Essays

Author: Isaac Persing ; Vincent Ng

Abstract: Recently, researchers have begun exploring methods of scoring student essays with respect to particular dimensions of quality such as coherence, technical errors, and relevance to prompt, but there is relatively little work on modeling thesis clarity. We present a new annotated corpus and propose a learning-based approach to scoring essays along the thesis clarity dimension. Additionally, in order to provide more valuable feedback on why an essay is scored as it is, we propose a second learning-based approach to identifying what kinds of errors an essay has that may lower its thesis clarity score.

6 0.061126862 368 acl-2013-Universal Dependency Annotation for Multilingual Parsing

7 0.060837314 58 acl-2013-Automated Collocation Suggestion for Japanese Second Language Learners

8 0.060256124 8 acl-2013-A Learner Corpus-based Approach to Verb Suggestion for ESL

9 0.059727531 369 acl-2013-Unsupervised Consonant-Vowel Prediction over Hundreds of Languages

10 0.055601653 323 acl-2013-Simpler unsupervised POS tagging with bilingual projections

11 0.054004986 139 acl-2013-Entity Linking for Tweets

12 0.051107991 184 acl-2013-Identification of Speakers in Novels

13 0.048870958 94 acl-2013-Coordination Structures in Dependency Treebanks

14 0.048548423 173 acl-2013-Graph-based Semi-Supervised Model for Joint Chinese Word Segmentation and Part-of-Speech Tagging

15 0.046015315 169 acl-2013-Generating Synthetic Comparable Questions for News Articles

16 0.044874005 235 acl-2013-Machine Translation Detection from Monolingual Web-Text

17 0.043753486 34 acl-2013-Accurate Word Segmentation using Transliteration and Language Model Projection

18 0.043392964 61 acl-2013-Automatic Interpretation of the English Possessive

19 0.043103233 276 acl-2013-Part-of-Speech Induction in Dependency Trees for Statistical Machine Translation

20 0.042830292 57 acl-2013-Arguments and Modifiers from the Learner's Perspective


similar papers computed by lsi model

lsi for this paper:

topicId topicWeight

[(0, 0.134), (1, -0.001), (2, -0.013), (3, -0.027), (4, -0.025), (5, -0.031), (6, -0.005), (7, 0.006), (8, 0.015), (9, -0.024), (10, 0.008), (11, 0.006), (12, 0.004), (13, -0.006), (14, -0.093), (15, -0.04), (16, 0.008), (17, -0.008), (18, -0.009), (19, 0.023), (20, 0.004), (21, 0.02), (22, 0.118), (23, -0.024), (24, 0.025), (25, 0.092), (26, 0.022), (27, -0.019), (28, 0.015), (29, -0.003), (30, -0.048), (31, -0.03), (32, -0.033), (33, -0.07), (34, -0.036), (35, 0.003), (36, -0.065), (37, 0.036), (38, 0.035), (39, -0.027), (40, -0.02), (41, 0.102), (42, -0.051), (43, -0.076), (44, 0.054), (45, 0.005), (46, -0.011), (47, 0.015), (48, -0.001), (49, -0.05)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.90494579 299 acl-2013-Reconstructing an Indo-European Family Tree from Non-native English Texts

Author: Ryo Nagata ; Edward Whittaker

Abstract: Mother tongue interference is the phenomenon where linguistic systems of a mother tongue are transferred to another language. Although there has been plenty of work on mother tongue interference, very little is known about how strongly it is transferred to another language and about what relation there is across mother tongues. To address these questions, this paper explores and visualizes mother tongue interference preserved in English texts written by Indo-European language speakers. This paper further explores linguistic features that explain why certain relations are preserved in English writing, and which contribute to related tasks such as native language identification.

2 0.63186389 246 acl-2013-Modeling Thesis Clarity in Student Essays

Author: Isaac Persing ; Vincent Ng

Abstract: Recently, researchers have begun exploring methods of scoring student essays with respect to particular dimensions of quality such as coherence, technical errors, and relevance to prompt, but there is relatively little work on modeling thesis clarity. We present a new annotated corpus and propose a learning-based approach to scoring essays along the thesis clarity dimension. Additionally, in order to provide more valuable feedback on why an essay is scored as it is, we propose a second learning-based approach to identifying what kinds of errors an essay has that may lower its thesis clarity score.

3 0.62812656 389 acl-2013-Word Association Profiles and their Use for Automated Scoring of Essays

Author: Beata Beigman Klebanov ; Michael Flor

Abstract: We describe a new representation of the content vocabulary of a text we call word association profile that captures the proportions of highly associated, mildly associated, unassociated, and dis-associated pairs of words that co-exist in the given text. We illustrate the shape of the distirbution and observe variation with genre and target audience. We present a study of the relationship between quality of writing and word association profiles. For a set of essays written by college graduates on a number of general topics, we show that the higher scoring essays tend to have higher percentages of both highly associated and dis-associated pairs, and lower percentages of mildly associated pairs of words. Finally, we use word association profiles to improve a system for automated scoring of essays.

4 0.58973461 122 acl-2013-Discriminative Approach to Fill-in-the-Blank Quiz Generation for Language Learners

Author: Keisuke Sakaguchi ; Yuki Arase ; Mamoru Komachi

Abstract: We propose discriminative methods to generate semantic distractors of fill-in-theblank quiz for language learners using a large-scale language learners’ corpus. Unlike previous studies, the proposed methods aim at satisfying both reliability and validity of generated distractors; distractors should be exclusive against answers to avoid multiple answers in one quiz, and distractors should discriminate learners’ proficiency. Detailed user evaluation with 3 native and 23 non-native speakers of English shows that our methods achieve better reliability and validity than previous methods.

5 0.56687373 58 acl-2013-Automated Collocation Suggestion for Japanese Second Language Learners

Author: Lis Pereira ; Erlyn Manguilimotan ; Yuji Matsumoto

Abstract: This study addresses issues of Japanese language learning concerning word combinations (collocations). Japanese learners may be able to construct grammatically correct sentences, however, these may sound “unnatural”. In this work, we analyze correct word combinations using different collocation measures and word similarity methods. While other methods use well-formed text, our approach makes use of a large Japanese language learner corpus for generating collocation candidates, in order to build a system that is more sensitive to constructions that are difficult for learners. Our results show that we get better results compared to other methods that use only wellformed text. 1

6 0.55381244 364 acl-2013-Typesetting for Improved Readability using Lexical and Syntactic Information

7 0.55299139 89 acl-2013-Computerized Analysis of a Verbal Fluency Test

8 0.55298811 8 acl-2013-A Learner Corpus-based Approach to Verb Suggestion for ESL

9 0.54299128 203 acl-2013-Is word-to-phone mapping better than phone-phone mapping for handling English words?

10 0.5309183 327 acl-2013-Sorani Kurdish versus Kurmanji Kurdish: An Empirical Comparison

11 0.52561843 186 acl-2013-Identifying English and Hungarian Light Verb Constructions: A Contrastive Approach

12 0.51177669 76 acl-2013-Building and Evaluating a Distributional Memory for Croatian

13 0.50789088 144 acl-2013-Explicit and Implicit Syntactic Features for Text Classification

14 0.50610703 102 acl-2013-DErivBase: Inducing and Evaluating a Derivational Morphology Resource for German

15 0.50248665 48 acl-2013-An Open Source Toolkit for Quantitative Historical Linguistics

16 0.49635714 227 acl-2013-Learning to lemmatise Polish noun phrases

17 0.48694986 149 acl-2013-Exploring Word Order Universals: a Probabilistic Graphical Model Approach

18 0.47325364 286 acl-2013-Psycholinguistically Motivated Computational Models on the Organization and Processing of Morphologically Complex Words

19 0.46490854 65 acl-2013-BRAINSUP: Brainstorming Support for Creative Sentence Generation

20 0.45914263 1 acl-2013-"Let Everything Turn Well in Your Wife": Generation of Adult Humor Using Lexical Constraints


similar papers computed by lda model

lda for this paper:

topicId topicWeight

[(0, 0.03), (6, 0.03), (11, 0.041), (15, 0.019), (24, 0.041), (25, 0.018), (26, 0.063), (35, 0.076), (42, 0.033), (48, 0.033), (70, 0.036), (88, 0.431), (90, 0.026), (95, 0.041)]

similar papers list:

simIndex simValue paperId paperTitle

1 0.94768113 106 acl-2013-Decentralized Entity-Level Modeling for Coreference Resolution

Author: Greg Durrett ; David Hall ; Dan Klein

Abstract: Efficiently incorporating entity-level information is a challenge for coreference resolution systems due to the difficulty of exact inference over partitions. We describe an end-to-end discriminative probabilistic model for coreference that, along with standard pairwise features, enforces structural agreement constraints between specified properties of coreferent mentions. This model can be represented as a factor graph for each document that admits efficient inference via belief propagation. We show that our method can use entity-level information to outperform a basic pairwise system.

2 0.92436498 141 acl-2013-Evaluating a City Exploration Dialogue System with Integrated Question-Answering and Pedestrian Navigation

Author: Srinivasan Janarthanam ; Oliver Lemon ; Phil Bartie ; Tiphaine Dalmas ; Anna Dickinson ; Xingkun Liu ; William Mackaness ; Bonnie Webber

Abstract: We present a city navigation and tourist information mobile dialogue app with integrated question-answering (QA) and geographic information system (GIS) modules that helps pedestrian users to navigate in and learn about urban environments. In contrast to existing mobile apps which treat these problems independently, our Android app addresses the problem of navigation and touristic questionanswering in an integrated fashion using a shared dialogue context. We evaluated our system in comparison with Samsung S-Voice (which interfaces to Google navigation and Google search) with 17 users and found that users judged our system to be significantly more interesting to interact with and learn from. They also rated our system above Google search (with the Samsung S-Voice interface) for tourist information tasks.

3 0.89939153 327 acl-2013-Sorani Kurdish versus Kurmanji Kurdish: An Empirical Comparison

Author: Kyumars Sheykh Esmaili ; Shahin Salavati

Abstract: Resource scarcity along with diversity– both in dialect and script–are the two primary challenges in Kurdish language processing. In this paper we aim at addressing these two problems by (i) building a text corpus for Sorani and Kurmanji, the two main dialects of Kurdish, and (ii) highlighting some of the orthographic, phonological, and morphological differences between these two dialects from statistical and rule-based perspectives.

same-paper 4 0.86965632 299 acl-2013-Reconstructing an Indo-European Family Tree from Non-native English Texts

Author: Ryo Nagata ; Edward Whittaker

Abstract: Mother tongue interference is the phenomenon where linguistic systems of a mother tongue are transferred to another language. Although there has been plenty of work on mother tongue interference, very little is known about how strongly it is transferred to another language and about what relation there is across mother tongues. To address these questions, this paper explores and visualizes mother tongue interference preserved in English texts written by Indo-European language speakers. This paper further explores linguistic features that explain why certain relations are preserved in English writing, and which contribute to related tasks such as native language identification.

5 0.8140617 136 acl-2013-Enhanced and Portable Dependency Projection Algorithms Using Interlinear Glossed Text

Author: Ryan Georgi ; Fei Xia ; William D. Lewis

Abstract: As most of the world’s languages are under-resourced, projection algorithms offer an enticing way to bootstrap the resources available for one resourcepoor language from a resource-rich language by means of parallel text and word alignment. These algorithms, however, make the strong assumption that the language pairs share common structures and that the parse trees will resemble one another. This assumption is useful but often leads to errors in projection. In this paper, we will address this weakness by using trees created from instances of Interlinear Glossed Text (IGT) to discover patterns of divergence between the lan- guages. We will show that this method improves the performance of projection algorithms significantly in some languages by accounting for divergence between languages using only the partial supervision of a few corrected trees.

6 0.7924487 41 acl-2013-Aggregated Word Pair Features for Implicit Discourse Relation Disambiguation

7 0.75973034 345 acl-2013-The Haves and the Have-Nots: Leveraging Unlabelled Corpora for Sentiment Analysis

8 0.64959854 111 acl-2013-Density Maximization in Context-Sense Metric Space for All-words WSD

9 0.61711693 252 acl-2013-Multigraph Clustering for Unsupervised Coreference Resolution

10 0.56836355 258 acl-2013-Neighbors Help: Bilingual Unsupervised WSD Using Context

11 0.50110728 105 acl-2013-DKPro WSD: A Generalized UIMA-based Framework for Word Sense Disambiguation

12 0.4988578 131 acl-2013-Dual Training and Dual Prediction for Polarity Classification

13 0.49556795 130 acl-2013-Domain-Specific Coreference Resolution with Lexicalized Features

14 0.49248117 382 acl-2013-Variational Inference for Structured NLP Models

15 0.48695421 196 acl-2013-Improving pairwise coreference models through feature space hierarchy learning

16 0.48353273 70 acl-2013-Bilingually-Guided Monolingual Dependency Grammar Induction

17 0.48335716 253 acl-2013-Multilingual Affect Polarity and Valence Prediction in Metaphor-Rich Texts

18 0.48040903 292 acl-2013-Question Classification Transfer

19 0.47669601 177 acl-2013-GuiTAR-based Pronominal Anaphora Resolution in Bengali

20 0.46480718 97 acl-2013-Cross-lingual Projections between Languages from Different Families