acl acl2011 acl2011-317 knowledge-graph by maker-knowledge-mining
Source: pdf
Author: Sina Zarriess ; Aoife Cahill ; Jonas Kuhn
Abstract: This paper addresses a data-driven surface realisation model based on a large-scale reversible grammar of German. We investigate the relationship between the surface realisation performance and the character of the input to generation, i.e. its degree of underspecification. We extend a syntactic surface realisation system, which can be trained to choose among word order variants, such that the candidate set includes active and passive variants. This allows us to study the interaction of voice and word order alternations in realistic German corpus data. We show that with an appropriately underspecified input, a linguistically informed realisation model trained to regenerate strings from the underlying semantic representation achieves 91.5% accuracy (over a baseline of 82.5%) in the prediction of the original voice. 1
Reference: text
sentIndex sentText sentNum sentScore
1 cahi l , fe l Abstract This paper addresses a data-driven surface realisation model based on a large-scale reversible grammar of German. [sent-3, score-0.832]
2 We investigate the relationship between the surface realisation performance and the character of the input to generation, i. [sent-4, score-0.733]
3 We extend a syntactic surface realisation system, which can be trained to choose among word order variants, such that the candidate set includes active and passive variants. [sent-7, score-1.183]
4 This allows us to study the interaction of voice and word order alternations in realistic German corpus data. [sent-8, score-0.48]
5 We show that with an appropriately underspecified input, a linguistically informed realisation model trained to regenerate strings from the underlying semantic representation achieves 91. [sent-9, score-0.848]
6 1 Introduction 1 This paper1 presents work on modelling the usage of voice and word order alternations in a free word order language. [sent-12, score-0.469]
7 Given a set of meaning-equivalent candidate sentences, such as in the simplified English Example (1), our model makes predictions about which candidate sentence is most appropriate or natural given the context. [sent-13, score-0.168]
8 It wasn’t until June that the Parliament approved it. [sent-16, score-0.049]
9 It wasn’t until June that it was approved by the Parliament. [sent-18, score-0.049]
10 Such ranking systems are practically relevant for the real-world applica- tion of grammar-based generators that usually generate several grammatical surface sentences from a given abstract input, e. [sent-26, score-0.298]
11 Moreover, this framework allows for detailed experimental studies of the interaction of specific linguistic features. [sent-29, score-0.064]
12 Thus it has been demonstrated that for free word order languages like German, word order prediction quality can be improved with carefully designed, linguistically informed models capturing information-structural strategies (Filippova and Strube, 2007; Cahill and Riester, 2009). [sent-30, score-0.193]
13 This paper is situated in the same framework, using rich linguistic representations over corpus data for machine learning of realisation ranking. [sent-31, score-0.631]
14 Quite obviously, word order is only one of the means at a speaker’s disposal for expressing some content in a contextually appropriate form; we add systematic alternations like the voice alternation (active vs. [sent-33, score-0.65]
15 As an alternative way of promoting or demoting the prominence of a syntactic argument, its interaction with word ordering strategies in real corpus data is of high theoretical interest (Aissen, 1999; Aissen, 2003; Bresnan et al. [sent-35, score-0.067]
16 We conduct a pilot human evaluation on the voice alProceedingPso orftla thned 4,9 Otrhe Agonnn,u Jauln Mee 1e9t-i2ng4, o 2f0 t1h1e. [sent-39, score-0.18]
17 c s 2o0ci1a1ti Aonss foocria Ctioomnp fourta Ctioomnaplu Ltaintigouniaslti Lcisn,g puaigsetsic 1s007–1017, ternation data and relate our findings to our results established in the automatic ranking experiments. [sent-41, score-0.111]
18 Addressing interactions among a range of grammatical and discourse phenomena on realistic corpus data turns out to be a major methodological challenge for data-driven surface realisation. [sent-42, score-0.35]
19 The set of candidate realisations available for ranking will influence the findings, and here, existing surface realisers vary considerably. [sent-43, score-0.469]
20 We study the effect of varying degrees of underspecification explicitly, extending a syntactic generation system by a semantic component capturing voice alternations. [sent-46, score-0.519]
21 In regeneration studies involving underspecified underlying representations, corpusoriented work reveals an additional methodological challenge. [sent-47, score-0.283]
22 When using standard semantic representations, as common in broad-coverage work in semantic parsing (i. [sent-48, score-0.06]
23 Rather than waiting for the availability of robust and reliable techniques for detecting the reference of implicit arguments in analysis (or for contextually aware reasoning components), we adopt a relatively simple heuristic approach (see Section 3. [sent-51, score-0.217]
24 1) that approximates the desired equivalences by augmented representations for examples like (1-c). [sent-52, score-0.129]
25 This way we can overcome an extremely skewed distribution in the naturally occurring meaning-equivalent active vs. [sent-53, score-0.119]
26 passive sentences, a factor which we believe justifies taking the risk of occasional overgeneration. [sent-54, score-0.291]
27 The paper is structured as follows: Section 2 situates our methodology with respect to other work on surface realisation and briefly summarises the relevant theoretical linguistic background. [sent-55, score-0.827]
28 In Section 3, we present our generation architecture and the design of the input representation. [sent-56, score-0.125]
29 1 Generation Background The first widely known data-driven approach to surface realisation, or tactical generation, (Langkilde and Knight, 1998) used language-model ngram statistics on a word lattice of candidate realisations to guide a ranker. [sent-60, score-0.389]
30 Subsequent work explored ways of exploiting linguistically annotated data for trainable generation models (Ratnaparkhi, 2000; Marciniak and Strube, 2005; Belz, 2005, a. [sent-61, score-0.139]
31 Work on data-driven approaches has led to insights into the importance of linguistic features for sentence linearisation decisions (Ringger et al. [sent-64, score-0.124]
32 The availability of discriminative learning techniques for the ranking of candidate analyses output by broad-coverage grammars with rich linguistic representations, originally in parsing (Riezler et al. [sent-66, score-0.18]
33 , 2002), has also led to a revival of interest in linguistically sophisticated reversible grammars as the basis for surface realisation (Velldal and Oepen, 2006; Cahill et al. [sent-68, score-0.855]
34 The grammar generates candidate analyses for an underlying representation and the ranker’s task is to predict the contextually appropriate realisation. [sent-70, score-0.361]
35 He uses an MRS representation derived by an HPSG grammar that can be underspecified for information status. [sent-72, score-0.271]
36 In his case, the underspecification is encoded in the grammar and not directly controlled. [sent-73, score-0.223]
37 (2010) generate from semantic corpus annotations included in the CoNLL’09 shared task data. [sent-75, score-0.03]
38 However, they note that these annotations are not suitable for full generation since they are often incomplete. [sent-76, score-0.092]
39 Thus, it is not clear to which degree these annotations are actually underspecified for certain paraphrases. [sent-77, score-0.183]
40 2 Linguistic Background In competition-based linguistic theories (Optimality Theory and related frameworks), the use of argument alternations is construed as an effect of markedness hierarchies (Aissen, 1999; Aissen, 2003). [sent-79, score-0.396]
41 ) on the one hand and the various properties that argument phrases can bear (person, animacy, definiteness) on the other are organised in markedness hierarchies. [sent-83, score-0.134]
42 Wherever possible, there is a tendency to align the hierarchies, i. [sent-84, score-0.031]
43 , use prominent functions to realise prominently marked argument phrases. [sent-86, score-0.07]
44 (2001) find that there is a statistical tendency in English to passivise a verb if the patient is higher on the person scale than the agent, but an active is grammatically possible. [sent-88, score-0.193]
45 (2007) correlate the use of the English dative alternation to a number of features such as givenness, pronominalisation, definiteness, constituent length, animacy of the involved verb arguments. [sent-90, score-0.172]
46 These features are assumed to reflect the discourse acessibility of the arguments. [sent-91, score-0.052]
47 Interestingly, the properties that have been used to model argument alternations in strict word order languages like English have been identified as factors that influence word order in free word order languages like German, see Filippova and Strube (2007) for a number of pointers. [sent-92, score-0.39]
48 Cahill and Riester (2009) implement a model for German word order variation that approximates the information status of constituents through morphological features like definiteness, pronominalisation etc. [sent-93, score-0.134]
49 We are not aware of any corpus-based generation studies investigating how these properties relate to argument alternations in free word order languages. [sent-94, score-0.531]
50 Two variants of this set-up that we use are sketched in Figure 1. [sent-98, score-0.031]
51 We generally use a hand-crafted, broad-coverage LFG for German (Rohrer and Forst, 2006) to parse a corpus sentence into a f(unctional) structure3 and generate all surface realisations from a given 2Compare the bidirectional competition set-up in some Optimality-Theoretic work, e. [sent-99, score-0.319]
52 Sntbn Figure 1: Generation pipelines f-structure, following the generation approach of Cahill et al. [sent-106, score-0.152]
53 F-structures are attributevalue matrices representing grammatical functions and morphosyntactic features; their theoretical motivation lies in the abstraction over details of surface realisation. [sent-108, score-0.251]
54 The grammar is implemented in the XLE framework (Crouch et al. [sent-109, score-0.054]
55 , 2006), which allows for reversible use of the same declarative grammar in the parsing and generation direction. [sent-110, score-0.224]
56 To obtain a more abstract underlying representation (in the pipeline on the right-hand side of Figure 1), the present work uses an additional semantic construction component (Crouch and King, 2006; Zarrieß, 2009) to map LFG f-structures to meaning representations. [sent-111, score-0.194]
57 For the reverse direction, the meaning representations are mapped to f-structures which can then be mapped to surface strings by the XLE generator (Zarrieß and Kuhn, 2010). [sent-112, score-0.412]
58 For the final realisation ranking step in both pipelines, we used SVMrank, a Support Vector Machine-based learning tool (Joachims, 1996). [sent-113, score-0.591]
59 The ranking step is thus technically independent from the LFG-based component. [sent-114, score-0.08]
60 However, the grammar is used to produce the training data, pairs of corpus sentences and the possible alternations. [sent-115, score-0.054]
61 The two pipelines allow us to vary the degree to which the generation input is underspecified. [sent-116, score-0.213]
62 the candidate set will contain just word order alternations. [sent-119, score-0.101]
63 In the semantic input, syntactic function and voice are underspecified, so a larger set of surface realisation candidates is generated. [sent-120, score-0.91]
64 Figure 2 illustrates the two representation levels for an active and a passive sentence. [sent-121, score-0.444]
65 The subject of the passive and the object of the active f-structure are mapped to the same role (patient) in the meaning representation. [sent-122, score-0.511]
66 1 Issues with “naive” underspecification In order to create an underspecified voice representation that does indeed leave open the realisation options available to the speaker/writer, it is often not sufficient to remove just the syntactic function information. [sent-124, score-1.108]
67 For instance, the subject of the active sentence (2) is an arbitrary reference pronoun man “one” which cannot be used as an oblique agent in a passive, sentence (2-b) is ungrammatical. [sent-125, score-0.414]
68 So, when combined with the grammar, the meaning representation for (2) in Figure 2 contains implicit information about the voice of the original corpus sentence; the candidate set will not include any passive realisations. [sent-132, score-0.711]
69 However, a passive realisation without the oblique agent in the by-phrase, as in Example (3), is a very natural variant. [sent-133, score-0.992]
70 The reverse situation arises frequently too: passive sentences where the agent role is not overtly realised. [sent-136, score-0.417]
71 Given the standard, “analysis-oriented” meaning representation for Sentence (4) in Figure 2, the realiser will not generate an active realisation since the agent role cannot be instantiated by any phrase in the grammar. [sent-137, score-0.961]
72 However, depending on the exact context there are typically options for realising the subject phrase in an active with very little descriptive content. [sent-138, score-0.152]
73 Ideally, one would like to account for these phenomena in a meaning representation that underspecifies the lexicalisation of discourse referents, and also captures the reference of implicit arguments. [sent-139, score-0.25]
74 In order to work around that problem, we implemented some simple heuristics which underspecify the realisation of certain verb arguments. [sent-141, score-0.542]
75 a set of pronouns (generic and neutral pronouns, universal quantifiers) that correspond to “trivial” agents in active and implicit agents 12-ro le tra n s . [sent-143, score-0.278]
76 71 %A%c(t8i(0v2%e%)18P0% as( 1(i26v% e) Table 1: Distribution of voices in SEMh (SEMn) in passive sentences; 2. [sent-144, score-0.327]
77 a set of prepositional adjuncts in passive sentences that correspond to subjects in active sentence (e. [sent-145, score-0.382]
78 certain syntactic contexts where special underspecification devices are needed, e. [sent-148, score-0.169]
79 In the following, we will distinguish 1-role transitives where the agent is “trivial” or implicit from 2-role transitives with a non-implicit agent. [sent-151, score-0.359]
80 By means ofthe extended underspecification rules for voice, the sentences in (2) and (3) receive an identical meaning representation. [sent-152, score-0.275]
81 As a result, our surface realiser can produce an active alternation for (3) and a passive alternation for (2). [sent-153, score-0.859]
82 In the following, we will refer to the extended representations as SEMh (“heuristic semantics”), and to the original representations as SEMn (“naive semantics”). [sent-154, score-0.18]
83 We are aware of the fact that these approximations introduce some noise into the data and do not always represent the underlying referents correctly. [sent-155, score-0.128]
84 For instance, the implicit agent in a passive need not be “trivial” but can correspond to an actual discourse referent. [sent-156, score-0.546]
85 However, we consider these heuristics as a first step towards capturing an important discourse function of the passive alternation, namely the deletion of the agent role. [sent-157, score-0.517]
86 If we did not treat the passives with an implicit agent on a par with certain actives, we would have to ignore a major portion of the passives occurring in corpus data. [sent-158, score-0.335]
87 Table 1 summarises the distribution of the voices for the heuristic meaning representation SEMh on the data-set we will introduce in Section 4, with the distribution for the naive representation SEMn in parentheses. [sent-159, score-0.352]
88 4 Experimental Set-up Data To obtain a sizable set of realistic corpus examples forour experiments onvoice alternations, we created our own dataset of input sentences and representations, instead of building on treebank examples as Cahill et al. [sent-160, score-0.077]
wordName wordTfidf (topN-words)
[('realisation', 0.511), ('passive', 0.263), ('alternations', 0.191), ('surface', 0.189), ('voice', 0.18), ('underspecification', 0.169), ('underspecified', 0.155), ('agent', 0.154), ('cahill', 0.146), ('kuhn', 0.133), ('realisations', 0.13), ('aissen', 0.128), ('active', 0.119), ('alternation', 0.116), ('zarrie', 0.113), ('contextually', 0.104), ('chancellor', 0.096), ('kanzler', 0.096), ('semh', 0.096), ('semn', 0.096), ('generation', 0.092), ('representations', 0.09), ('strube', 0.088), ('ranking', 0.08), ('velldal', 0.078), ('filippova', 0.078), ('bresnan', 0.078), ('reversible', 0.078), ('implicit', 0.077), ('wasn', 0.073), ('candidate', 0.07), ('german', 0.07), ('argument', 0.07), ('definiteness', 0.069), ('linearisation', 0.064), ('markedness', 0.064), ('oblique', 0.064), ('pronominalisation', 0.064), ('riester', 0.064), ('rohrer', 0.064), ('summarises', 0.064), ('transitives', 0.064), ('voices', 0.064), ('wurde', 0.064), ('representation', 0.062), ('pipelines', 0.06), ('meaning', 0.059), ('xle', 0.056), ('realiser', 0.056), ('animacy', 0.056), ('forst', 0.056), ('grammar', 0.054), ('discourse', 0.052), ('passives', 0.052), ('june', 0.049), ('referents', 0.049), ('regeneration', 0.049), ('approved', 0.049), ('belz', 0.049), ('capturing', 0.048), ('linguistically', 0.047), ('receive', 0.047), ('trivial', 0.047), ('crouch', 0.046), ('ranker', 0.046), ('jonas', 0.046), ('oepen', 0.046), ('realistic', 0.044), ('man', 0.044), ('parliament', 0.044), ('lfg', 0.044), ('riezler', 0.044), ('investigating', 0.044), ('underlying', 0.043), ('patient', 0.043), ('agents', 0.041), ('hierarchies', 0.041), ('naive', 0.041), ('approximates', 0.039), ('mapped', 0.037), ('free', 0.036), ('aware', 0.036), ('methodological', 0.036), ('interaction', 0.034), ('input', 0.033), ('theoretical', 0.033), ('subject', 0.033), ('der', 0.032), ('variants', 0.031), ('tendency', 0.031), ('order', 0.031), ('relate', 0.031), ('led', 0.03), ('linguistic', 0.03), ('semantic', 0.03), ('goals', 0.029), ('grammatical', 0.029), ('appropriate', 0.028), ('degree', 0.028), ('occasional', 0.028)]
simIndex simValue paperId paperTitle
same-paper 1 0.99999976 317 acl-2011-Underspecifying and Predicting Voice for Surface Realisation Ranking
Author: Sina Zarriess ; Aoife Cahill ; Jonas Kuhn
Abstract: This paper addresses a data-driven surface realisation model based on a large-scale reversible grammar of German. We investigate the relationship between the surface realisation performance and the character of the input to generation, i.e. its degree of underspecification. We extend a syntactic surface realisation system, which can be trained to choose among word order variants, such that the candidate set includes active and passive variants. This allows us to study the interaction of voice and word order alternations in realistic German corpus data. We show that with an appropriately underspecified input, a linguistically informed realisation model trained to regenerate strings from the underlying semantic representation achieves 91.5% accuracy (over a baseline of 82.5%) in the prediction of the original voice. 1
Author: Nina Dethlefs ; Heriberto Cuayahuitl
Abstract: Surface realisation decisions in language generation can be sensitive to a language model, but also to decisions of content selection. We therefore propose the joint optimisation of content selection and surface realisation using Hierarchical Reinforcement Learning (HRL). To this end, we suggest a novel reward function that is induced from human data and is especially suited for surface realisation. It is based on a generation space in the form of a Hidden Markov Model (HMM). Results in terms of task success and human-likeness suggest that our unified approach performs better than greedy or random baselines.
3 0.12611899 267 acl-2011-Reversible Stochastic Attribute-Value Grammars
Author: Daniel de Kok ; Barbara Plank ; Gertjan van Noord
Abstract: An attractive property of attribute-value grammars is their reversibility. Attribute-value grammars are usually coupled with separate statistical components for parse selection and fluency ranking. We propose reversible stochastic attribute-value grammars, in which a single statistical model is employed both for parse selection and fluency ranking.
4 0.09442021 219 acl-2011-Metagrammar engineering: Towards systematic exploration of implemented grammars
Author: Antske Fokkens
Abstract: When designing grammars of natural language, typically, more than one formal analysis can account for a given phenomenon. Moreover, because analyses interact, the choices made by the engineer influence the possibilities available in further grammar development. The order in which phenomena are treated may therefore have a major impact on the resulting grammar. This paper proposes to tackle this problem by using metagrammar development as a methodology for grammar engineering. Iargue that metagrammar engineering as an approach facilitates the systematic exploration of grammars through comparison of competing analyses. The idea is illustrated through a comparative study of auxiliary structures in HPSG-based grammars for German and Dutch. Auxiliaries form a central phenomenon of German and Dutch and are likely to influence many components of the grammar. This study shows that a special auxiliary+verb construction significantly improves efficiency compared to the standard argument-composition analysis for both parsing and generation.
5 0.083573863 128 acl-2011-Exploring Entity Relations for Named Entity Disambiguation
Author: Danuta Ploch
Abstract: Named entity disambiguation is the task of linking an entity mention in a text to the correct real-world referent predefined in a knowledge base, and is a crucial subtask in many areas like information retrieval or topic detection and tracking. Named entity disambiguation is challenging because entity mentions can be ambiguous and an entity can be referenced by different surface forms. We present an approach that exploits Wikipedia relations between entities co-occurring with the ambiguous form to derive a range of novel features for classifying candidate referents. We find that our features improve disambiguation results significantly over a strong popularity baseline, and are especially suitable for recognizing entities not contained in the knowledge base. Our system achieves state-of-the-art results on the TAC-KBP 2009 dataset.
6 0.062587857 324 acl-2011-Unsupervised Semantic Role Induction via Split-Merge Clustering
7 0.059841909 157 acl-2011-I Thou Thee, Thou Traitor: Predicting Formal vs. Informal Address in English Literature
8 0.059329819 53 acl-2011-Automatically Evaluating Text Coherence Using Discourse Relations
9 0.057892174 3 acl-2011-A Bayesian Model for Unsupervised Semantic Parsing
10 0.051458512 298 acl-2011-The ACL Anthology Searchbench
11 0.051304437 236 acl-2011-Optimistic Backtracking - A Backtracking Overlay for Deterministic Incremental Parsing
12 0.051064003 119 acl-2011-Evaluating the Impact of Coder Errors on Active Learning
13 0.050041895 238 acl-2011-P11-2093 k2opt.pdf
14 0.049061429 202 acl-2011-Learning Hierarchical Translation Structure with Linguistic Annotations
15 0.048966594 79 acl-2011-Confidence Driven Unsupervised Semantic Parsing
16 0.048396908 110 acl-2011-Effective Use of Function Words for Rule Generalization in Forest-Based Translation
17 0.04772003 269 acl-2011-Scaling up Automatic Cross-Lingual Semantic Role Annotation
18 0.045138538 164 acl-2011-Improving Arabic Dependency Parsing with Form-based and Functional Morphological Features
19 0.043893777 322 acl-2011-Unsupervised Learning of Semantic Relation Composition
20 0.043236516 11 acl-2011-A Fast and Accurate Method for Approximate String Search
topicId topicWeight
[(0, 0.132), (1, -0.002), (2, -0.048), (3, -0.033), (4, -0.045), (5, 0.04), (6, 0.018), (7, 0.007), (8, -0.026), (9, -0.02), (10, -0.021), (11, -0.036), (12, -0.016), (13, 0.003), (14, -0.017), (15, -0.023), (16, -0.011), (17, -0.008), (18, -0.023), (19, 0.002), (20, 0.052), (21, 0.043), (22, -0.056), (23, -0.115), (24, -0.042), (25, 0.032), (26, -0.035), (27, -0.02), (28, -0.04), (29, -0.102), (30, 0.038), (31, 0.127), (32, 0.043), (33, -0.009), (34, 0.038), (35, 0.174), (36, -0.223), (37, -0.036), (38, 0.028), (39, -0.05), (40, 0.046), (41, -0.112), (42, -0.317), (43, 0.106), (44, -0.102), (45, -0.02), (46, -0.09), (47, 0.027), (48, 0.041), (49, 0.183)]
simIndex simValue paperId paperTitle
same-paper 1 0.94534278 317 acl-2011-Underspecifying and Predicting Voice for Surface Realisation Ranking
Author: Sina Zarriess ; Aoife Cahill ; Jonas Kuhn
Abstract: This paper addresses a data-driven surface realisation model based on a large-scale reversible grammar of German. We investigate the relationship between the surface realisation performance and the character of the input to generation, i.e. its degree of underspecification. We extend a syntactic surface realisation system, which can be trained to choose among word order variants, such that the candidate set includes active and passive variants. This allows us to study the interaction of voice and word order alternations in realistic German corpus data. We show that with an appropriately underspecified input, a linguistically informed realisation model trained to regenerate strings from the underlying semantic representation achieves 91.5% accuracy (over a baseline of 82.5%) in the prediction of the original voice. 1
Author: Nina Dethlefs ; Heriberto Cuayahuitl
Abstract: Surface realisation decisions in language generation can be sensitive to a language model, but also to decisions of content selection. We therefore propose the joint optimisation of content selection and surface realisation using Hierarchical Reinforcement Learning (HRL). To this end, we suggest a novel reward function that is induced from human data and is especially suited for surface realisation. It is based on a generation space in the form of a Hidden Markov Model (HMM). Results in terms of task success and human-likeness suggest that our unified approach performs better than greedy or random baselines.
3 0.5701071 219 acl-2011-Metagrammar engineering: Towards systematic exploration of implemented grammars
Author: Antske Fokkens
Abstract: When designing grammars of natural language, typically, more than one formal analysis can account for a given phenomenon. Moreover, because analyses interact, the choices made by the engineer influence the possibilities available in further grammar development. The order in which phenomena are treated may therefore have a major impact on the resulting grammar. This paper proposes to tackle this problem by using metagrammar development as a methodology for grammar engineering. Iargue that metagrammar engineering as an approach facilitates the systematic exploration of grammars through comparison of competing analyses. The idea is illustrated through a comparative study of auxiliary structures in HPSG-based grammars for German and Dutch. Auxiliaries form a central phenomenon of German and Dutch and are likely to influence many components of the grammar. This study shows that a special auxiliary+verb construction significantly improves efficiency compared to the standard argument-composition analysis for both parsing and generation.
4 0.56122011 267 acl-2011-Reversible Stochastic Attribute-Value Grammars
Author: Daniel de Kok ; Barbara Plank ; Gertjan van Noord
Abstract: An attractive property of attribute-value grammars is their reversibility. Attribute-value grammars are usually coupled with separate statistical components for parse selection and fluency ranking. We propose reversible stochastic attribute-value grammars, in which a single statistical model is employed both for parse selection and fluency ranking.
5 0.51635039 102 acl-2011-Does Size Matter - How Much Data is Required to Train a REG Algorithm?
Author: Mariet Theune ; Ruud Koolen ; Emiel Krahmer ; Sander Wubben
Abstract: In this paper we investigate how much data is required to train an algorithm for attribute selection, a subtask of Referring Expressions Generation (REG). To enable comparison between different-sized training sets, a systematic training method was developed. The results show that depending on the complexity of the domain, training on 10 to 20 items may already lead to a good performance.
7 0.46481368 68 acl-2011-Classifying arguments by scheme
8 0.46262628 252 acl-2011-Prototyping virtual instructors from human-human corpora
9 0.41860855 303 acl-2011-Tier-based Strictly Local Constraints for Phonology
10 0.40928513 42 acl-2011-An Interface for Rapid Natural Language Processing Development in UIMA
11 0.40465263 239 acl-2011-P11-5002 k2opt.pdf
12 0.37996683 321 acl-2011-Unsupervised Discovery of Rhyme Schemes
13 0.36267594 236 acl-2011-Optimistic Backtracking - A Backtracking Overlay for Deterministic Incremental Parsing
14 0.36191851 138 acl-2011-French TimeBank: An ISO-TimeML Annotated Reference Corpus
15 0.35667172 298 acl-2011-The ACL Anthology Searchbench
16 0.33370936 119 acl-2011-Evaluating the Impact of Coder Errors on Active Learning
17 0.33068329 128 acl-2011-Exploring Entity Relations for Named Entity Disambiguation
18 0.32943258 322 acl-2011-Unsupervised Learning of Semantic Relation Composition
19 0.32051268 125 acl-2011-Exploiting Readymades in Linguistic Creativity: A System Demonstration of the Jigsaw Bard
20 0.31608409 320 acl-2011-Unsupervised Discovery of Domain-Specific Knowledge from Text
topicId topicWeight
[(5, 0.031), (17, 0.047), (31, 0.432), (37, 0.078), (39, 0.033), (41, 0.064), (55, 0.027), (59, 0.064), (72, 0.023), (91, 0.022), (96, 0.084)]
simIndex simValue paperId paperTitle
1 0.8258388 211 acl-2011-Liars and Saviors in a Sentiment Annotated Corpus of Comments to Political Debates
Author: Paula Carvalho ; Luis Sarmento ; Jorge Teixeira ; Mario J. Silva
Abstract: We investigate the expression of opinions about human entities in user-generated content (UGC). A set of 2,800 online news comments (8,000 sentences) was manually annotated, following a rich annotation scheme designed for this purpose. We conclude that the challenge in performing opinion mining in such type of content is correctly identifying the positive opinions, because (i) they are much less frequent than negative opinions and (ii) they are particularly exposed to verbal irony. We also show that the recognition of human targets poses additional challenges on mining opinions from UGC, since they are frequently mentioned by pronouns, definite descriptions and nicknames. 1
same-paper 2 0.79330254 317 acl-2011-Underspecifying and Predicting Voice for Surface Realisation Ranking
Author: Sina Zarriess ; Aoife Cahill ; Jonas Kuhn
Abstract: This paper addresses a data-driven surface realisation model based on a large-scale reversible grammar of German. We investigate the relationship between the surface realisation performance and the character of the input to generation, i.e. its degree of underspecification. We extend a syntactic surface realisation system, which can be trained to choose among word order variants, such that the candidate set includes active and passive variants. This allows us to study the interaction of voice and word order alternations in realistic German corpus data. We show that with an appropriately underspecified input, a linguistically informed realisation model trained to regenerate strings from the underlying semantic representation achieves 91.5% accuracy (over a baseline of 82.5%) in the prediction of the original voice. 1
3 0.57870442 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
4 0.57803327 301 acl-2011-The impact of language models and loss functions on repair disfluency detection
Author: Simon Zwarts ; Mark Johnson
Abstract: Unrehearsed spoken language often contains disfluencies. In order to correctly interpret a spoken utterance, any such disfluencies must be identified and removed or otherwise dealt with. Operating on transcripts of speech which contain disfluencies, we study the effect of language model and loss function on the performance of a linear reranker that rescores the 25-best output of a noisychannel model. We show that language models trained on large amounts of non-speech data improve performance more than a language model trained on a more modest amount of speech data, and that optimising f-score rather than log loss improves disfluency detection performance. Our approach uses a log-linear reranker, operating on the top n analyses of a noisy channel model. We use large language models, introduce new features into this reranker and . examine different optimisation strategies. We obtain a disfluency detection f-scores of 0.838 which improves upon the current state-of-theart.
Author: Chi-kiu Lo ; Dekai Wu
Abstract: We introduce a novel semi-automated metric, MEANT, that assesses translation utility by matching semantic role fillers, producing scores that correlate with human judgment as well as HTER but at much lower labor cost. As machine translation systems improve in lexical choice and fluency, the shortcomings of widespread n-gram based, fluency-oriented MT evaluation metrics such as BLEU, which fail to properly evaluate adequacy, become more apparent. But more accurate, nonautomatic adequacy-oriented MT evaluation metrics like HTER are highly labor-intensive, which bottlenecks the evaluation cycle. We first show that when using untrained monolingual readers to annotate semantic roles in MT output, the non-automatic version of the metric HMEANT achieves a 0.43 correlation coefficient with human adequacyjudgments at the sentence level, far superior to BLEU at only 0.20, and equal to the far more expensive HTER. We then replace the human semantic role annotators with automatic shallow semantic parsing to further automate the evaluation metric, and show that even the semiautomated evaluation metric achieves a 0.34 correlation coefficient with human adequacy judgment, which is still about 80% as closely correlated as HTER despite an even lower labor cost for the evaluation procedure. The results show that our proposed metric is significantly better correlated with human judgment on adequacy than current widespread automatic evaluation metrics, while being much more cost effective than HTER. 1
6 0.42818442 308 acl-2011-Towards a Framework for Abstractive Summarization of Multimodal Documents
7 0.40661365 32 acl-2011-Algorithm Selection and Model Adaptation for ESL Correction Tasks
8 0.39593726 274 acl-2011-Semi-Supervised Frame-Semantic Parsing for Unknown Predicates
9 0.38850889 84 acl-2011-Contrasting Opposing Views of News Articles on Contentious Issues
10 0.38540444 164 acl-2011-Improving Arabic Dependency Parsing with Form-based and Functional Morphological Features
11 0.37882257 128 acl-2011-Exploring Entity Relations for Named Entity Disambiguation
12 0.36938882 289 acl-2011-Subjectivity and Sentiment Analysis of Modern Standard Arabic
13 0.35980141 149 acl-2011-Hierarchical Reinforcement Learning and Hidden Markov Models for Task-Oriented Natural Language Generation
14 0.35685441 7 acl-2011-A Corpus for Modeling Morpho-Syntactic Agreement in Arabic: Gender, Number and Rationality
15 0.35427386 3 acl-2011-A Bayesian Model for Unsupervised Semantic Parsing
16 0.35357839 273 acl-2011-Semantic Representation of Negation Using Focus Detection
17 0.35295981 311 acl-2011-Translationese and Its Dialects
18 0.34840816 292 acl-2011-Target-dependent Twitter Sentiment Classification
19 0.34831467 267 acl-2011-Reversible Stochastic Attribute-Value Grammars
20 0.34779063 286 acl-2011-Social Network Extraction from Texts: A Thesis Proposal