acl acl2013 acl2013-180 knowledge-graph by maker-knowledge-mining

180 acl-2013-Handling Ambiguities of Bilingual Predicate-Argument Structures for Statistical Machine Translation


Source: pdf

Author: Feifei Zhai ; Jiajun Zhang ; Yu Zhou ; Chengqing Zong

Abstract: Predicate-argument structure (PAS) has been demonstrated to be very effective in improving SMT performance. However, since a sourceside PAS might correspond to multiple different target-side PASs, there usually exist many PAS ambiguities during translation. In this paper, we group PAS ambiguities into two types: role ambiguity and gap ambiguity. Then we propose two novel methods to handle the two PAS ambiguities for SMT accordingly: 1) inside context integration; 2) a novel maximum entropy PAS disambiguation (MEPD) model. In this way, we incorporate rich context information of PAS for disambiguation. Then we integrate the two methods into a PASbased translation framework. Experiments show that our approach helps to achieve significant improvements on translation quality. 1

Reference: text


Summary: the most important sentenses genereted by tfidf model

sentIndex sentText sentNum sentScore

1 cn Abstract Predicate-argument structure (PAS) has been demonstrated to be very effective in improving SMT performance. [sent-4, score-0.022]

2 However, since a sourceside PAS might correspond to multiple different target-side PASs, there usually exist many PAS ambiguities during translation. [sent-5, score-0.258]

3 In this paper, we group PAS ambiguities into two types: role ambiguity and gap ambiguity. [sent-6, score-0.395]

4 Then we propose two novel methods to handle the two PAS ambiguities for SMT accordingly: 1) inside context integration; 2) a novel maximum entropy PAS disambiguation (MEPD) model. [sent-7, score-0.384]

5 In this way, we incorporate rich context information of PAS for disambiguation. [sent-8, score-0.076]

6 Then we integrate the two methods into a PASbased translation framework. [sent-9, score-0.089]

7 Experiments show that our approach helps to achieve significant improvements on translation quality. [sent-10, score-0.066]

8 1 Introduction Predicate-argument structure (PAS) depicts the relationship between a predicate and its associated arguments, which indicates the skeleton structure of a sentence on semantic level. [sent-11, score-0.133]

9 Basically, PAS agrees much better between two languages than syntax structure (Fung et al. [sent-12, score-0.047]

10 Considering that current syntaxbased translation models are always impaired by cross-lingual structure divergence (Eisner, 2003; Zhang et al. [sent-14, score-0.166]

11 , 2010), PAS is really a better representation of a sentence pair to model the bilingual structure mapping. [sent-15, score-0.07]

12 However, since a source-side PAS might correspond to multiple different target-side PASs, there usually exist many PAS ambiguities during translation. [sent-16, score-0.229]

13 For example, in Figure 1, (a) and (b) carry the same source-side PAS <[A0]1 [Pred(是)]2 [A1]3> for Chinese predicate “是”. [sent-17, score-0.044]

14 However, in Figure 1(a), the corresponding target-side-like PAS is <[X1] [X2] [X3]>, while in Figure 1(b), the counterpart target-side-like PAS 1 is <[X2] [X3] [X1]>. [sent-18, score-0.022]

15 This is because the two PASs play different roles in their corresponding sentences. [sent-19, score-0.026]

16 Actually, Figure 1(a) is an independ- ent PAS, while Figure 1(b) is a modifier of the noun phrase “中 国 和 俄罗斯”. [sent-20, score-0.036]

17 We call this kind of PAS ambiguity role ambiguity. [sent-21, score-0.169]

18 Meanwhile, Figure 1 also depicts another kind of PAS ambiguity. [sent-24, score-0.065]

19 However, they are different because in Figure 1(c), there is a gap string “对 运动员” between [A0] and [Pred]. [sent-26, score-0.139]

20 Generally, the gap strings are due to the low recall of automatic semantic role labeling (SRL) or complex sentence structures. [sent-27, score-0.185]

21 For example, in Figure 1(c), the gap string “对 运动员 ” is actually an argument “AM-PRP” of the PAS, but the SRL system has 1We use target-side-like PAS to refer to a list of general non-terminals in target language order, where a nonterminal aligns to a source argument. [sent-28, score-0.203]

22 1127 Proce dingsS o f ita h,e B 5u1lgsta Arinan,u Aaulg Musete 4ti-n9g 2 o0f1 t3h. [sent-29, score-0.013]

23 Ac s2s0o1ci3a Atiosnso fcoirat Cio nm foprut Caotimonpaulta Lti nognuails Lti cnsg,u piasgteics 1 27–1 36, ignored it. [sent-31, score-0.03]

24 We call this kind of PAS ambiguity gap ambiguity. [sent-32, score-0.252]

25 During translation, these PAS ambiguities will greatly affect the PAS-based translation models. [sent-33, score-0.244]

26 Therefore, in order to incorporate the bilingual PAS into machine translation effectively, we need to decide which target-side-like PAS should be chosen for a specific source-side PAS. [sent-34, score-0.126]

27 In this paper, we propose two novel methods to incorporate rich context information to handle PAS ambiguities. [sent-36, score-0.132]

28 Towards the gap ambiguity, we adopt a method called inside context integration to extend PAS to IC-PAS. [sent-37, score-0.237]

29 In terms of IC-PAS, the gap strings are combined effectively to deal with the gap ambiguities. [sent-38, score-0.29]

30 As to the role ambiguity, we design a novel maximum entropy PAS disambiguation (MEPD) model to combine various context features, such as context words of PAS. [sent-39, score-0.2]

31 For each ambiguous source-side PAS, we build a specific MEPD model to select appropriate target-side-like PAS for translation. [sent-40, score-0.024]

32 We will detail the two methods in Section 3 and 4 respectively. [sent-41, score-0.012]

33 Finally, we integrate the above two methods into a PAS-based translation framework (Zhai et al. [sent-42, score-0.105]

34 Experiments show that the two PAS disambiguation methods significantly improve the baseline translation system. [sent-44, score-0.108]

35 The main contribution of this work can be concluded as follows: 1) We define two kinds of PAS ambiguities: role ambiguity and gap ambiguity. [sent-45, score-0.256]

36 To our best knowledge, we are the first to handle these PAS ambiguities for SMT. [sent-46, score-0.193]

37 2) Towards the two different ambiguities, we design two specific methods for PAS disambiguation: inside context integration and the novel MEPD model. [sent-47, score-0.163]

38 2 PAS-based Translation Framework PAS-based translation framework is to perform translation based on PAS transformation (Zhai et al. [sent-48, score-0.21]

39 In the framework, a source-side PAS is first converted into target-side-like PASs by PAS transformation rules, and then perform translation based on the obtained target-side-like PASs. [sent-50, score-0.141]

40 1 PAS Transformation Rules PAS transformation rules (PASTR) are used to convert a source-side PAS into a target one. [sent-52, score-0.095]


similar papers computed by tfidf model

tfidf for this paper:

wordName wordTfidf (topN-words)

[('pas', 0.913), ('ambiguities', 0.164), ('mepd', 0.156), ('pred', 0.123), ('gap', 0.118), ('ambiguity', 0.078), ('pastr', 0.078), ('zhai', 0.078), ('pass', 0.071), ('translation', 0.066), ('transformation', 0.062), ('inside', 0.045), ('integration', 0.045), ('disambiguation', 0.042), ('srl', 0.04), ('fung', 0.04), ('depicts', 0.036), ('role', 0.035), ('yzhou', 0.035), ('impaired', 0.032), ('athletes', 0.032), ('feifei', 0.032), ('strings', 0.032), ('bilingual', 0.03), ('incorporate', 0.03), ('context', 0.029), ('handle', 0.029), ('kind', 0.029), ('syntaxbased', 0.029), ('olympic', 0.029), ('sourceside', 0.029), ('predicate', 0.028), ('russia', 0.027), ('jiajun', 0.027), ('call', 0.027), ('novel', 0.027), ('nlpr', 0.025), ('skeleton', 0.025), ('agrees', 0.025), ('ong', 0.025), ('concluded', 0.025), ('chengqing', 0.024), ('zong', 0.024), ('smt', 0.024), ('ambiguous', 0.024), ('zhang', 0.023), ('integrate', 0.023), ('countries', 0.023), ('effectively', 0.022), ('ff', 0.022), ('nonterminal', 0.022), ('counterpart', 0.022), ('structure', 0.022), ('china', 0.022), ('automation', 0.021), ('aligns', 0.021), ('entropy', 0.021), ('lti', 0.021), ('string', 0.021), ('actually', 0.021), ('exist', 0.02), ('rules', 0.019), ('correspond', 0.019), ('meanwhile', 0.019), ('triple', 0.019), ('ent', 0.019), ('eisner', 0.019), ('chinese', 0.018), ('ignored', 0.018), ('really', 0.018), ('basically', 0.018), ('design', 0.017), ('divergence', 0.017), ('modifier', 0.017), ('accordingly', 0.017), ('rich', 0.017), ('framework', 0.016), ('carry', 0.016), ('academy', 0.015), ('yu', 0.014), ('greatly', 0.014), ('convert', 0.014), ('figure', 0.014), ('primary', 0.014), ('zhou', 0.014), ('handling', 0.014), ('play', 0.013), ('roles', 0.013), ('towards', 0.013), ('converted', 0.013), ('location', 0.013), ('beijing', 0.013), ('cio', 0.013), ('dingss', 0.013), ('might', 0.013), ('usually', 0.013), ('ita', 0.013), ('detail', 0.012), ('laboratory', 0.012), ('nm', 0.012)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 1.0000001 180 acl-2013-Handling Ambiguities of Bilingual Predicate-Argument Structures for Statistical Machine Translation

Author: Feifei Zhai ; Jiajun Zhang ; Yu Zhou ; Chengqing Zong

Abstract: Predicate-argument structure (PAS) has been demonstrated to be very effective in improving SMT performance. However, since a sourceside PAS might correspond to multiple different target-side PASs, there usually exist many PAS ambiguities during translation. In this paper, we group PAS ambiguities into two types: role ambiguity and gap ambiguity. Then we propose two novel methods to handle the two PAS ambiguities for SMT accordingly: 1) inside context integration; 2) a novel maximum entropy PAS disambiguation (MEPD) model. In this way, we incorporate rich context information of PAS for disambiguation. Then we integrate the two methods into a PASbased translation framework. Experiments show that our approach helps to achieve significant improvements on translation quality. 1

2 0.097347751 200 acl-2013-Integrating Phrase-based Reordering Features into a Chart-based Decoder for Machine Translation

Author: ThuyLinh Nguyen ; Stephan Vogel

Abstract: Hiero translation models have two limitations compared to phrase-based models: 1) Limited hypothesis space; 2) No lexicalized reordering model. We propose an extension of Hiero called PhrasalHiero to address Hiero’s second problem. Phrasal-Hiero still has the same hypothesis space as the original Hiero but incorporates a phrase-based distance cost feature and lexicalized reodering features into the chart decoder. The work consists of two parts: 1) for each Hiero translation derivation, find its corresponding dis- continuous phrase-based path. 2) Extend the chart decoder to incorporate features from the phrase-based path. We achieve significant improvement over both Hiero and phrase-based baselines for ArabicEnglish, Chinese-English and GermanEnglish translation.

3 0.092721045 373 acl-2013-Using Conceptual Class Attributes to Characterize Social Media Users

Author: Shane Bergsma ; Benjamin Van Durme

Abstract: We describe a novel approach for automatically predicting the hidden demographic properties of social media users. Building on prior work in common-sense knowledge acquisition from third-person text, we first learn the distinguishing attributes of certain classes of people. For example, we learn that people in the Female class tend to have maiden names and engagement rings. We then show that this knowledge can be used in the analysis of first-person communication; knowledge of distinguishing attributes allows us to both classify users and to bootstrap new training examples. Our novel approach enables substantial improvements on the widelystudied task of user gender prediction, ob- taining a 20% relative error reduction over the current state-of-the-art.

4 0.059522048 314 acl-2013-Semantic Roles for String to Tree Machine Translation

Author: Marzieh Bazrafshan ; Daniel Gildea

Abstract: We experiment with adding semantic role information to a string-to-tree machine translation system based on the rule extraction procedure of Galley et al. (2004). We compare methods based on augmenting the set of nonterminals by adding semantic role labels, and altering the rule extraction process to produce a separate set of rules for each predicate that encompass its entire predicate-argument structure. Our results demonstrate that the second approach is effective in increasing the quality of translations.

5 0.055706069 223 acl-2013-Learning a Phrase-based Translation Model from Monolingual Data with Application to Domain Adaptation

Author: Jiajun Zhang ; Chengqing Zong

Abstract: Currently, almost all of the statistical machine translation (SMT) models are trained with the parallel corpora in some specific domains. However, when it comes to a language pair or a different domain without any bilingual resources, the traditional SMT loses its power. Recently, some research works study the unsupervised SMT for inducing a simple word-based translation model from the monolingual corpora. It successfully bypasses the constraint of bitext for SMT and obtains a relatively promising result. In this paper, we take a step forward and propose a simple but effective method to induce a phrase-based model from the monolingual corpora given an automatically-induced translation lexicon or a manually-edited translation dictionary. We apply our method for the domain adaptation task and the extensive experiments show that our proposed method can substantially improve the translation quality. 1

6 0.054483328 62 acl-2013-Automatic Term Ambiguity Detection

7 0.040610503 291 acl-2013-Question Answering Using Enhanced Lexical Semantic Models

8 0.037386037 359 acl-2013-Translating Dialectal Arabic to English

9 0.035126641 228 acl-2013-Leveraging Domain-Independent Information in Semantic Parsing

10 0.035065159 98 acl-2013-Cross-lingual Transfer of Semantic Role Labeling Models

11 0.034303591 10 acl-2013-A Markov Model of Machine Translation using Non-parametric Bayesian Inference

12 0.031200139 93 acl-2013-Context Vector Disambiguation for Bilingual Lexicon Extraction from Comparable Corpora

13 0.030357754 19 acl-2013-A Shift-Reduce Parsing Algorithm for Phrase-based String-to-Dependency Translation

14 0.030327551 63 acl-2013-Automatic detection of deception in child-produced speech using syntactic complexity features

15 0.029352859 16 acl-2013-A Novel Translation Framework Based on Rhetorical Structure Theory

16 0.029170191 329 acl-2013-Statistical Machine Translation Improves Question Retrieval in Community Question Answering via Matrix Factorization

17 0.028918238 68 acl-2013-Bilingual Data Cleaning for SMT using Graph-based Random Walk

18 0.028854059 255 acl-2013-Name-aware Machine Translation

19 0.028462946 181 acl-2013-Hierarchical Phrase Table Combination for Machine Translation

20 0.02835332 39 acl-2013-Addressing Ambiguity in Unsupervised Part-of-Speech Induction with Substitute Vectors


similar papers computed by lsi model

lsi for this paper:

topicId topicWeight

[(0, 0.068), (1, -0.036), (2, 0.031), (3, 0.005), (4, 0.002), (5, 0.003), (6, -0.023), (7, 0.01), (8, 0.033), (9, 0.028), (10, 0.009), (11, 0.02), (12, 0.025), (13, 0.018), (14, 0.025), (15, -0.004), (16, 0.032), (17, -0.007), (18, 0.008), (19, 0.009), (20, -0.018), (21, -0.001), (22, -0.018), (23, -0.035), (24, 0.026), (25, -0.01), (26, 0.03), (27, -0.013), (28, 0.037), (29, 0.003), (30, -0.016), (31, -0.005), (32, 0.019), (33, -0.022), (34, 0.004), (35, 0.045), (36, -0.02), (37, -0.031), (38, -0.03), (39, -0.011), (40, -0.021), (41, 0.071), (42, 0.005), (43, 0.032), (44, -0.024), (45, -0.039), (46, 0.038), (47, 0.017), (48, 0.012), (49, -0.024)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.89557064 180 acl-2013-Handling Ambiguities of Bilingual Predicate-Argument Structures for Statistical Machine Translation

Author: Feifei Zhai ; Jiajun Zhang ; Yu Zhou ; Chengqing Zong

Abstract: Predicate-argument structure (PAS) has been demonstrated to be very effective in improving SMT performance. However, since a sourceside PAS might correspond to multiple different target-side PASs, there usually exist many PAS ambiguities during translation. In this paper, we group PAS ambiguities into two types: role ambiguity and gap ambiguity. Then we propose two novel methods to handle the two PAS ambiguities for SMT accordingly: 1) inside context integration; 2) a novel maximum entropy PAS disambiguation (MEPD) model. In this way, we incorporate rich context information of PAS for disambiguation. Then we integrate the two methods into a PASbased translation framework. Experiments show that our approach helps to achieve significant improvements on translation quality. 1

2 0.57298982 314 acl-2013-Semantic Roles for String to Tree Machine Translation

Author: Marzieh Bazrafshan ; Daniel Gildea

Abstract: We experiment with adding semantic role information to a string-to-tree machine translation system based on the rule extraction procedure of Galley et al. (2004). We compare methods based on augmenting the set of nonterminals by adding semantic role labels, and altering the rule extraction process to produce a separate set of rules for each predicate that encompass its entire predicate-argument structure. Our results demonstrate that the second approach is effective in increasing the quality of translations.

3 0.5199337 320 acl-2013-Shallow Local Multi-Bottom-up Tree Transducers in Statistical Machine Translation

Author: Fabienne Braune ; Nina Seemann ; Daniel Quernheim ; Andreas Maletti

Abstract: We present a new translation model integrating the shallow local multi bottomup tree transducer. We perform a largescale empirical evaluation of our obtained system, which demonstrates that we significantly beat a realistic tree-to-tree baseline on the WMT 2009 English → German tlriannes olnati tohne tWasMk.T TA 2s0 an a Edndgitliisonha →l c Gonetrrmibauntion we make the developed software and complete tool-chain publicly available for further experimentation.

4 0.51503891 16 acl-2013-A Novel Translation Framework Based on Rhetorical Structure Theory

Author: Mei Tu ; Yu Zhou ; Chengqing Zong

Abstract: Rhetorical structure theory (RST) is widely used for discourse understanding, which represents a discourse as a hierarchically semantic structure. In this paper, we propose a novel translation framework with the help of RST. In our framework, the translation process mainly includes three steps: 1) Source RST-tree acquisition: a source sentence is parsed into an RST tree; 2) Rule extraction: translation rules are extracted from the source tree and the target string via bilingual word alignment; 3) RST-based translation: the source RST-tree is translated with translation rules. Experiments on Chinese-to-English show that our RST-based approach achieves improvements of 2.3/0.77/1.43 BLEU points on NIST04/NIST05/CWMT2008 respectively. 1

5 0.51251978 10 acl-2013-A Markov Model of Machine Translation using Non-parametric Bayesian Inference

Author: Yang Feng ; Trevor Cohn

Abstract: Most modern machine translation systems use phrase pairs as translation units, allowing for accurate modelling of phraseinternal translation and reordering. However phrase-based approaches are much less able to model sentence level effects between different phrase-pairs. We propose a new model to address this imbalance, based on a word-based Markov model of translation which generates target translations left-to-right. Our model encodes word and phrase level phenomena by conditioning translation decisions on previous decisions and uses a hierarchical Pitman-Yor Process prior to provide dynamic adaptive smoothing. This mechanism implicitly supports not only traditional phrase pairs, but also gapping phrases which are non-consecutive in the source. Our experiments on Chinese to English and Arabic to English translation show consistent improvements over competitive baselines, of up to +3.4 BLEU.

6 0.50901884 361 acl-2013-Travatar: A Forest-to-String Machine Translation Engine based on Tree Transducers

7 0.49641326 330 acl-2013-Stem Translation with Affix-Based Rule Selection for Agglutinative Languages

8 0.49635312 11 acl-2013-A Multi-Domain Translation Model Framework for Statistical Machine Translation

9 0.47873393 200 acl-2013-Integrating Phrase-based Reordering Features into a Chart-based Decoder for Machine Translation

10 0.4746885 38 acl-2013-Additive Neural Networks for Statistical Machine Translation

11 0.4741835 255 acl-2013-Name-aware Machine Translation

12 0.47324869 92 acl-2013-Context-Dependent Multilingual Lexical Lookup for Under-Resourced Languages

13 0.47193485 360 acl-2013-Translating Italian connectives into Italian Sign Language

14 0.46711913 378 acl-2013-Using subcategorization knowledge to improve case prediction for translation to German

15 0.46520734 312 acl-2013-Semantic Parsing as Machine Translation

16 0.45810929 223 acl-2013-Learning a Phrase-based Translation Model from Monolingual Data with Application to Domain Adaptation

17 0.44471815 226 acl-2013-Learning to Prune: Context-Sensitive Pruning for Syntactic MT

18 0.44451964 154 acl-2013-Extracting bilingual terminologies from comparable corpora

19 0.44385558 201 acl-2013-Integrating Translation Memory into Phrase-Based Machine Translation during Decoding

20 0.43023995 110 acl-2013-Deepfix: Statistical Post-editing of Statistical Machine Translation Using Deep Syntactic Analysis


similar papers computed by lda model

lda for this paper:

topicId topicWeight

[(0, 0.043), (6, 0.012), (11, 0.025), (15, 0.031), (19, 0.026), (24, 0.053), (26, 0.031), (35, 0.046), (42, 0.056), (48, 0.07), (68, 0.283), (70, 0.071), (88, 0.024), (90, 0.02), (95, 0.058)]

similar papers list:

simIndex simValue paperId paperTitle

same-paper 1 0.79508072 180 acl-2013-Handling Ambiguities of Bilingual Predicate-Argument Structures for Statistical Machine Translation

Author: Feifei Zhai ; Jiajun Zhang ; Yu Zhou ; Chengqing Zong

Abstract: Predicate-argument structure (PAS) has been demonstrated to be very effective in improving SMT performance. However, since a sourceside PAS might correspond to multiple different target-side PASs, there usually exist many PAS ambiguities during translation. In this paper, we group PAS ambiguities into two types: role ambiguity and gap ambiguity. Then we propose two novel methods to handle the two PAS ambiguities for SMT accordingly: 1) inside context integration; 2) a novel maximum entropy PAS disambiguation (MEPD) model. In this way, we incorporate rich context information of PAS for disambiguation. Then we integrate the two methods into a PASbased translation framework. Experiments show that our approach helps to achieve significant improvements on translation quality. 1

2 0.70474917 51 acl-2013-AnnoMarket: An Open Cloud Platform for NLP

Author: Valentin Tablan ; Kalina Bontcheva ; Ian Roberts ; Hamish Cunningham ; Marin Dimitrov

Abstract: This paper presents AnnoMarket, an open cloud-based platform which enables researchers to deploy, share, and use language processing components and resources, following the data-as-a-service and software-as-a-service paradigms. The focus is on multilingual text analysis resources and services, based on an opensource infrastructure and compliant with relevant NLP standards. We demonstrate how the AnnoMarket platform can be used to develop NLP applications with little or no programming, to index the results for enhanced browsing and search, and to evaluate performance. Utilising AnnoMarket is straightforward, since cloud infrastructural issues are dealt with by the platform, completely transparently to the user: load balancing, efficient data upload and storage, deployment on the virtual machines, security, and fault tolerance.

3 0.48090681 326 acl-2013-Social Text Normalization using Contextual Graph Random Walks

Author: Hany Hassan ; Arul Menezes

Abstract: We introduce a social media text normalization system that can be deployed as a preprocessing step for Machine Translation and various NLP applications to handle social media text. The proposed system is based on unsupervised learning of the normalization equivalences from unlabeled text. The proposed approach uses Random Walks on a contextual similarity bipartite graph constructed from n-gram sequences on large unlabeled text corpus. We show that the proposed approach has a very high precision of (92.43) and a reasonable recall of (56.4). When used as a preprocessing step for a state-of-the-art machine translation system, the translation quality on social media text improved by 6%. The proposed approach is domain and language independent and can be deployed as a preprocessing step for any NLP application to handle social media text.

4 0.4663215 82 acl-2013-Co-regularizing character-based and word-based models for semi-supervised Chinese word segmentation

Author: Xiaodong Zeng ; Derek F. Wong ; Lidia S. Chao ; Isabel Trancoso

Abstract: This paper presents a semi-supervised Chinese word segmentation (CWS) approach that co-regularizes character-based and word-based models. Similarly to multi-view learning, the “segmentation agreements” between the two different types of view are used to overcome the scarcity of the label information on unlabeled data. The proposed approach trains a character-based and word-based model on labeled data, respectively, as the initial models. Then, the two models are constantly updated using unlabeled examples, where the learning objective is maximizing their segmentation agreements. The agreements are regarded as a set of valuable constraints for regularizing the learning of both models on unlabeled data. The segmentation for an input sentence is decoded by using a joint scoring function combining the two induced models. The evaluation on the Chinese tree bank reveals that our model results in better gains over the state-of-the-art semi-supervised models reported in the literature.

5 0.45865816 249 acl-2013-Models of Semantic Representation with Visual Attributes

Author: Carina Silberer ; Vittorio Ferrari ; Mirella Lapata

Abstract: We consider the problem of grounding the meaning of words in the physical world and focus on the visual modality which we represent by visual attributes. We create a new large-scale taxonomy of visual attributes covering more than 500 concepts and their corresponding 688K images. We use this dataset to train attribute classifiers and integrate their predictions with text-based distributional models of word meaning. We show that these bimodal models give a better fit to human word association data compared to amodal models and word representations based on handcrafted norming data.

6 0.45683527 329 acl-2013-Statistical Machine Translation Improves Question Retrieval in Community Question Answering via Matrix Factorization

7 0.45677555 80 acl-2013-Chinese Parsing Exploiting Characters

8 0.45620638 164 acl-2013-FudanNLP: A Toolkit for Chinese Natural Language Processing

9 0.45222932 275 acl-2013-Parsing with Compositional Vector Grammars

10 0.45157173 47 acl-2013-An Information Theoretic Approach to Bilingual Word Clustering

11 0.45140269 222 acl-2013-Learning Semantic Textual Similarity with Structural Representations

12 0.44817066 224 acl-2013-Learning to Extract International Relations from Political Context

13 0.44705912 187 acl-2013-Identifying Opinion Subgroups in Arabic Online Discussions

14 0.44696715 123 acl-2013-Discriminative Learning with Natural Annotations: Word Segmentation as a Case Study

15 0.44669494 78 acl-2013-Categorization of Turkish News Documents with Morphological Analysis

16 0.4465003 56 acl-2013-Argument Inference from Relevant Event Mentions in Chinese Argument Extraction

17 0.44602054 276 acl-2013-Part-of-Speech Induction in Dependency Trees for Statistical Machine Translation

18 0.44562697 173 acl-2013-Graph-based Semi-Supervised Model for Joint Chinese Word Segmentation and Part-of-Speech Tagging

19 0.44529855 254 acl-2013-Multimodal DBN for Predicting High-Quality Answers in cQA portals

20 0.44505087 233 acl-2013-Linking Tweets to News: A Framework to Enrich Short Text Data in Social Media