A multilingual FrameNet-based grammar and lexicon for controlled natural language

Berkeley FrameNet is a lexico-semantic resource for English based on the theory of frame semantics. It has been exploited in a range of natural language processing applications and has inspired the development of framenets for many languages. We present a methodological approach to the extraction and generation of a computational multilingual FrameNet-based grammar and lexicon. The approach leverages FrameNet-annotated corpora to automatically extract a set of cross-lingual semantico-syntactic valence patterns. Based on data from Berkeley FrameNet and Swedish FrameNet, the proposed approach has been implemented in Grammatical Framework (GF), a categorial grammar formalism specialized for multilingual grammars. The implementation of the grammar and lexicon is supported by the design of FrameNet, providing a frame semantic abstraction layer, an interlingual semantic application programming interface (API), over the interlingual syntactic API already provided by GF Resource Grammar Library. The evaluation of the acquired grammar and lexicon shows the feasibility of the approach. Additionally, we illustrate how the FrameNet-based grammar and lexicon are exploited in two distinct multilingual controlled natural language applications. The produced resources are available under an open source license.     

A frame-based approach for capturing semantics from Arabic text for text-to-sign language MT

This paper describes the design and implementation of a computational model for Arabic natural language semantics, a semantic parser for capturing the deep semantic representation of Arabic text. The parser represents a major part of an Interlingua-based machine translation system for translating Arabic text into Sign Language. The parser follows a frame-based analysis to capture the overall meaning of Arabic text into a formal representation suitable for NLP applications that need for deep semantics representation, such as language generation and machine translation. We will show the representational power of this theory for the semantic analysis of texts in Arabic, a language which differs substantially from English in several ways. We will also show that the integration of WordNet and FrameNet in a single unified knowledge resource can improve disambiguation accuracy. Furthermore, we will propose a rule based algorithm to generate an equivalent Arabic FrameNet, using a lexical resource alignment of FrameNet1.3 LUs and WordNet3.0 synsets for English Language. A pilot study of motion and location verbs was carried out in order to test our system. Our corpus is made up of more than 2000 Arabic sentences in the domain of motion events collected from Algerian first level educational Arabic books and other relevant Arabic corpora.     

Machine translation from Arabic to English and French

As the cognitive processes of natural language understanding and generation are better understood, it is becoming easier, nowadays, to perform machine translation. In this paper we present our work on machine translation from Arabic to English and French, and illustrate it with a fully operational system, which runs on PC compatibles with Arabic/Latin interface. This system is an extension of an earlier system, whose task was the analysis of the natural language Arabic. Thanks to the regularity of its phrase structures and word patterns, Arabic lends itself quite naturally to a Fillmore-like analysis. The meaning of a phrase is stored in a star-like data structure, where the verb occupies the center of the star and the various noun sentences occupy specific peripheral nodes of the star. The data structure is then translated into an internal representation in the target language, which is then mapped into the target text.     

The NESPOLE! System for multilingual speech communication over the Internet

The NESPOLE! System is a speech communication system designed to support multilingual interaction between common users and providers of e-commerce services over the Internet. The core of the system is a distributed interlingua-based speech-to-speech translation system, which is supported by multimodal capabilities that allow the two parties participating in the communication to share Web pages and graphical content which can be annotated using gestures. We describe the unique features and considerations behind the design and implementation of this system, and evaluate these within the context of a constructed full prototype of the system that was developed for the domain of travel planning.     

英汉《小王子》抽象语义图结构的对比分析

AMR(抽象语义表示)是国际上一种新的句子语义表示方法,有着接近于中间语言的表示能力,其研发者已经建立了英文《小王子》等AMR语料库。AMR与以往的句法语义表示方法的最大不同在于两个方面,首先采用图结构来表示句子的语义;其次允许添加原句之外的概念节点来表示隐含的语义。该文针对汉语特点,在制定中文AMR标注规范的基础上,标注完成了中文版《小王子》的AMR语料库,标注一致性的Smatch值为0.83。统计结果显示,英汉双语含图结构句子具有很高的相关性,且含有图的句子比例高达40%左右,额外添加的概念节点则存在较大差异。最后讨论了AMR在汉语句子语义表示以及跨语言对比方面的优势。     

Symbolic-to-statistical hybridization: extending generation-heavy machine translation

The last few years have witnessed an increasing interest in hybridizing surface-based statistical approaches and rule-based symbolic approaches to machine translation (MT). Much of that work is focused on extending statistical MT systems with symbolic knowledge and components. In the brand of hybridization discussed here, we go in the opposite direction: adding statistical bilingual components to a symbolic system. Our base system is Generation-heavy machine translation (GHMT), a primarily symbolic asymmetrical approach that addresses the issue of Interlingual MT resource poverty in source-poor/target-rich language pairs by exploiting symbolic and statistical target-language resources. GHMT’s statistical components are limited to target-language models, which arguably makes it a simple form of a hybrid system. We extend the hybrid nature of GHMT by adding statistical bilingual components. We also describe the details of retargeting it to Arabic–English MT. The morphological richness of Arabic brings several challenges to the hybridization task. We conduct an extensive evaluation of multiple system variants. Our evaluation shows that this new variant of GHMT—a primarily symbolic system extended with monolingual and bilingual statistical components—has a higher degree of grammaticality than a phrase-based statistical MT system, where grammaticality is measured in terms of correct verb-argument realization and long-distance dependency translation.     

Parsing Arabic using induced probabilistic context free grammar

The importance of the parsing task for NLP applications is well understood. However developing parsers remains difficult because of the complexity of the Arabic language. Most parsers are based on syntactic grammars that describe the syntactic structures of a language. The development of these grammars is laborious and time consuming. In this paper we present our method for building an Arabic parser based on an induced grammar, PCFG grammar. We first induce the PCFG grammar from an Arabic Treebank. Then, we implement the parser that assigns syntactic structure to each input sentence. The parser is tested on sentences extracted from the treebank (1650 sentences).We calculate the precision, recall and f-measure. Our experimental results showed the efficiency of the proposed parser for parsing modern standard Arabic sentences (Precision: 83.59 %, Recall: 82.98 % and F-measure: 83.23 %).     

Matxin, an open-source rule-based machine translation system for Basque

We present the first publicly available machine translation (MT) system for Basque. The fact that Basque is both a morphologically rich and less-resourced language makes the use of statistical approaches difficult, and raises the need to develop a rule-based architecture which can be combined in the future with statistical techniques. The MT architecture proposed reuses several open-source tools and is based on a unique XML format to facilitate the flow between the different modules, which eases the interaction among different developers of tools and resources. The result is the rule-based Matxin MT system, an open-source toolkit, whose first implementation translates from Spanish to Basque. We have performed innovative work on the following tasks: construction of a dependency analyser for Spanish, use of rich linguistic information to translate prepositions and syntactic functions (such as subject and object markers), construction of an efficient module for verbal chunk transfer, and design and implementation of modules for ordering words and phrases, independently of the source language.     

基于源语言句法增强解码的神经机器翻译方法

当前性能最优的机器翻译模型之一Transformer基于标准的端到端结构,仅依赖于平行句对,默认模型能够自动学习语料中的知识;但这种建模方式缺乏显式的引导,不能有效挖掘深层语言知识,特别是在语料规模和质量受限的低资源环境下,句子解码缺乏先验约束,从而造成译文质量下降。为了缓解上述问题,提出了基于源语言句法增强解码的神经机器翻译（SSED）方法,显式地引入源语句句法信息指导解码。所提方法首先利用源语句句法信息构造句法感知的遮挡机制,引导编码自注意力生成一个额外的句法相关表征;然后将句法相关表征作为原句表征的补充,通过注意力机制融入解码,共同指导目标语言的生成,实现对模型的先验句法增强。在多个IWSLT及WMT标准机器翻译评测任务测试集上的实验结果显示,与Transformer基线模型相比,所提方法的BLEU值提高了0.84~3.41,达到了句法相关研究的最先进水平。句法信息与自注意力机制融合是有效的,利用源语言句法可指导神经机器翻译系统的解码过程,显著提高译文质量。     

Using target-language information to train part-of-speech taggers for machine translation

Although corpus-based approaches to machine translation (MT) are growing in interest, they are not applicable when the translation involves less-resourced language pairs for which there are no parallel corpora available; in those cases, the rule-based approach is the only applicable solution. Most rule-based MT systems make use of part-of-speech (PoS) taggers to solve the PoS ambiguities in the source-language texts to translate; those MT systems require accurate PoS taggers to produce reliable translations in the target language (TL). The standard statistical approach to PoS ambiguity resolution (or tagging) uses hidden Markov models (HMM) trained in a supervised way from hand-tagged corpora, an expensive resource not always available, or in an unsupervised way through the Baum-Welch expectation-maximization algorithm; both methods use information only from the language being tagged. However, when tagging is considered as an intermediate task for the translation procedure, that is, when the PoS tagger is to be embedded as a module within an MT system, information from the TL can be (unsupervisedly) used in the training phase to increase the translation quality of the whole MT system. This paper presents a method to train HMM-based PoS taggers to be used in MT; the new method uses not only information from the source language (SL), as general-purpose methods do, but also information from the TL and from the remaining modules of the MT system in which the PoS tagger is to be embedded. We find that the translation quality of the MT system embedding a PoS tagger trained in an unsupervised manner through this new method is clearly better than that of the same MT system embedding a PoS tagger trained through the Baum-Welch algorithm, and comparable to that obtained by embedding a PoS tagger trained in a supervised way from hand-tagged corpora.     

Joint Morphological-Lexical Language Modeling for Processing Morphologically Rich Languages With Application to Dialectal Arabic

Language modeling for an inflected language such as Arabic poses new challenges for speech recognition and machine translation due to its rich morphology. Rich morphology results in large increases in out-of-vocabulary (OOV) rate and poor language model parameter estimation in the absence of large quantities of data. In this study, we present a joint morphological-lexical language model (JMLLM) that takes advantage of Arabic morphology. JMLLM combines morphological segments with the underlying lexical items and additional available information sources with regards to morphological segments and lexical items in a single joint model. Joint representation and modeling of morphological and lexical items reduces the OOV rate and provides smooth probability estimates while keeping the predictive power of whole words. Speech recognition and machine translation experiments in dialectal-Arabic show improvements over word and morpheme based trigram language models. We also show that as the tightness of integration between different information sources increases, both speech recognition and machine translation performances improve.      

Exploiting Wikipedia and EuroWordNet to solve Cross-Lingual Question Answering

This paper describes a new advance in solving Cross-Lingual Question Answering (CL-QA) tasks. It is built on three main pillars: (i) the use of several multilingual knowledge resources to reference words between languages (the Inter Lingual Index (ILI) module of EuroWordNet and the multilingual knowledge encoded in Wikipedia); (ii) the consideration of more than only one translation per word in order to search candidate answers; and (iii) the analysis of the question in the original language without any translation process. This novel approach overcomes the errors caused by the common use of Machine Translation (MT) services by CL-QA systems. We also expose some studies and experiments that justify the importance of analyzing whether a Named Entity should be translated or not. Experimental results in bilingual scenarios show that our approach performs better than an MT based CL-QA approach achieving an average improvement of 36.7%.     

The potential of Swetra — A multilanguage MT system

One may indicate the potentials of an MT system by stating what text genres it can process, e.g., weather reports and technical manuals. This approach is practical, but misleading, unless domain knowledge is highly integrated in the system. Another way to indicate which fragments of language the system can process is to state its grammatical potentials, or more formally, which languages the grammars of the system can generate. This approach is more technical and less understandable to the layman (customer), but it is less misleading, since it stresses the point that the fragments which can be translated by the grammars of a system need not necessarily coincide exactly with any particular genre. Generally, the syntactic and lexical rules of an MT system allow it to translate many sentences other than those belonging to a certain genre. On the other hand it probably cannot translate all the sentences of a particular genre. Swetra is a multilanguage MT system defined by the potentials of a formal grammar (standard referent grammar) and not by reference to a genre. Successful translation of sentences can be guaranteed if they are within a specified syntactic format based on a specified lexicon. The paper discusses the consequences of this approach (Grammatically Restricted Machine Translation, GRMT) and describes the limits set by a standard choice of grammatical rules for sentences and clauses, noun phrases, verb phrases, sentence adverbials, etc. Such rules have been set up for English, Swedish and Russian, mainly on the basis of familiarity (frequency) and computer efficiency, but restricting the grammar and making it suitable for several languages poses many problems for optimization. Sample texts — newspaper reports — illustrate the type of text that can be translated with reasonable success among Russian, English and Swedish.     

MARS: A Statistical Semantic Parsing and Generation-Based Multilingual Automatic tRanslation System

We present MARS (Multilingual Automatic tRanslation System), a research prototype speech-to-speech translation system. MARS is aimed at two-way conversational spoken language translation between English and Mandarin Chinese for limited domains, such as air travel reservations. In MARS, machine translation is embedded within a complex speech processing task, and the translation performance is highly effected by the performance of other components, such as the recognizer and semantic parser, etc. All components in the proposed system are statistically trained using an appropriate training corpus. The speech signal is first recognized by an automatic speech recognizer (ASR). Next, the ASR-transcribed text is analyzed by a semantic parser, which uses a statistical decision-tree model that does not require hand-crafted grammars or rules. Furthermore, the parser provides semantic information that helps further re-scoring of the speech recognition hypotheses. The semantic content extracted by the parser is formatted into a language-independent tree structure, which is used for an interlingua based translation. A Maximum Entropy based sentence-level natural language generation (NLG) approach is used to generate sentences in the target language from the semantic tree representations. Finally, the generated target sentence is synthesized into speech by a speech synthesizer.Many new features and innovations have been incorporated into MARS: the translation is based on understanding the meaning of the sentence; the semantic parser uses a statistical model and is trained from a semantically annotated corpus; the output of the semantic parser is used to select a more specific language model to refine the speech recognition performance; the NLG component uses a statistical model and is also trained from the same annotated corpus. These features give MARS the advantages of robustness to speech disfluencies and recognition errors, tighter integration of semantic information into speech recognition, and portability to new languages and domains. These advantages are verified by our experimental results.     

Improving phrase-based statistical machine translation with morphosyntactic transformation

We present a phrase-based statistical machine translation approach which uses linguistic analysis in the preprocessing phase. The linguistic analysis includes morphological transformation and syntactic transformation. Since the word-order problem is solved using syntactic transformation, there is no reordering in the decoding phase. For morphological transformation, we use hand-crafted transformational rules. For syntactic transformation, we propose a transformational model based on a probabilistic context-free grammar. This model is trained using a bilingual corpus and a broad-coverage parser of the source language. This approach is applicable to language pairs in which the target language is poor in resources. We considered translation from English to Vietnamese and from English to French. Our experiments showed significant BLEU-score improvements in comparison with Pharaoh, a state-of-the-art phrase-based SMT system.     

Interlingua-based English–Hindi Machine Translation and Language Divergence

Interlingua and transfer-based approaches tomachine translation have long been in use in competing and complementary ways. The former proves economical in situations where translation among multiple languages is involved, and can be used as a knowledge-representation scheme. But given a particular interlingua, its adoption depends on its ability (a) to capture the knowledge in texts precisely and accurately and (b) to handle cross-language divergences. This paper studies the language divergence between English and Hindi and its implication to machine translation between these languages using the Universal Networking Language (UNL). UNL has been introduced by the United Nations University, Tokyo, to facilitate the transfer and exchange of information over the internet. The representation works at the level of single sentences and defines a semantic net-like structure in which nodes are word concepts and arcs are semantic relations between these concepts. The language divergences between Hindi, an Indo-European language, and English can be considered as representing the divergences between the SOV and SVO classes of languages. The work presented here is the only one to our knowledge that describes language divergence phenomena in the framework of computational linguistics through a South Asian language.     

From syntactic encodings to thematic roles: Building lexical entries for interlingual MT

Our goal is to construct large-scale lexicons for interlingual MT of English, Arabic, Korean, and Spanish. We describe techniques that predict salient linguistic features of a non-English word using the features of its English gloss (i.e., translation) in a bilingual dictionary. While not exact, owing to inexact glosses and language-to-language variations, these techniques can augment an existing dictionary with reasonable accuracy, thus saving significant time. We have conducted two experiments that demonstrate the value of these techniques. The first tested the feasibility of building a database of thematic grids for over 6500 Arabic verbs based on a mapping between English glosses and the syntactic codes in Longman's Dictionary of Contemporary English (LDOCE) (Procter, 1978). We show that it is more efficient and less error-prone to hand-verify the automatically constructed grids than it would be to build the thematic grids by hand from scratch. The second experiment tested the automatic classification of verbs into a richer semantic typology based on (Levin, 1993), from which we can derive a more refined set of thematic grids. In this second experiment, we show that a brute-force, non-robust technique provides 72% accuracy for semantic classification of LDOCE verbs; we then show that it is possible to approach this yield with a more robust technique based on fine-tuned statistical correlations. We further suggest the possibility of raising this yield by taking into account linguistic factors such as polysemy and positive and negative constraints on the syntax-semantics relation. We conclude that, while human intervention will always be necessary for the construction of a semantic classification from LDOCE, such intervention is significantly minimized as more knowledge about the syntax-semantics relation is introduced.     

Glossary-Based MT Engines in a Multilingual Analyst's Workstation Architecture

Fast access to information in different languages is still a major problem for many organizations. We have built a multilingual analyst's workstation integrated in the Tipster document management toolkit. The analyst workstation offers to an English-speaking analyst a variety of tools to browse sets of documents in Arabic, Japanese, Spanish and Russian, including a Unicode-based multilingual editor, and a simple machine translation functionality.The Temple project has developed an open multilingual architecture and software support for rapid development of extensible machine translation functionalities. The targeted languages are those for which natural language processing and human resources are scarce or difficult to obtain. The goal is to support rapid development of machine translation functionalities in a very short time with limited resources.Glossary-based machine-translation (GBMT) is used to provide an English gloss of a foreign document. A GBMT system uses a bilingual phrasal dictionary (glossary) to produce a phrase-by-phrase translation. Translation (based on phrase pattern-matching) is fast and accurate regarding the content of the document and browsed documents can be translated almost in real-time. A GBMT system for a language pair is also extremely simple, cheap and fast to develop. Moreover, all language resources used by the system are entirely under the control of the user.     

The potential of swetra—A multilanguage MT system

One may indicate the potentials of an MT system by stating what text genres it can process, e.g., weather reports and technical manuals. This approach is practical, but misleading, unless domain knowledge is highly integrated in the system. Another way to indicate which fragments of language the system can process is to state its grammatical potentials, or more formally, which languages the grammars of the system can generate. This approach is more technical and less understandable to the layman (customer), but it is less misleading, since it stresses the point that the fragments which can be translated by the grammars of a system need not necessarily coincide exactly with any particular genre. Generally, the syntactic and lexical rules of an MT system allow it to translate many sentences other than those belonging to a certain genre. On the other hand it probably cannot translate all the sentences of a particular genre. Swetra is a multilanguage MT system defined by the potentials of a formal grammar (standard referent grammar) and not by reference to a genre. Successful translation of sentences can be guaranteed if they are within a specified syntactic format based on a specified lexicon. The paper discusses the consequences of this approach (Grammatically Restricted Machine Translation, GRMT) and describes the limits set by a standard choice of grammatical rules for sentences and clauses, noun phrases, verb phrases, sentence adverbials, etc. Such rules have been set up for English, Swedish and Russian, mainly on the basis of familiarity (frequency) and computer efficiency, but restricting the grammar and making it suitable for several languages poses many problems for optimization. Sample texts—newspaper reports—illustrate the type of text that can be translated with reasonable success among Russian, English and Swedish.     

Multilingual natural language generation for multilingual software: A functional linguistic approach

In this paper we present an implemented account of multilingual linguistic resources for multilingual text generation that improves significantly on the degree of reuse of resources both across languages and across applications. We argue that this is a necessary step for multilingual generation in order to reduce the high cost of constructing linguistic resources and to make natural language generation relevant for a wider range of applications particularly, in this paper, for multilingual software and user interfaces. We begin by contrasting a weak and a strong approach to multilinguality in the state of the art in multilingual text generation. Neither approach has provided sufficient principles for organizing multilingual work. We then introduce our framework , where multilingual variation is included as an intrinsic feature of all levels of representation. We provide an example of multilingual tactical generation using this approach and discuss some of the performance, maintenance, and development issues that arise.