Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
18625 papers
Understanding emotional reactions to robots is becoming increasingly more important as technological capabilities expand. Just as computers have evolved, so have robots. This evolution has lead to significant applications across a wide variety of social contexts. This study examines the affective judgment of multiple robotic platforms across various contexts. Significant differences were found in both positive and negative affect ratings. To further explain the results, future research is proposed to investigate the specific physical robotic characteristics that elicit both positive and negative emotional response.
The aim of this journal article is to measure the productivity of two Old English verbal suffixes that display a low index of productivity, namely -ettan and -laecan. The main sources of this research are the lexical database of Old English Nert- hus and the Dictionary of Old English Corpus. Methodologically, two indexes of pro- ductivity are measured, productivity in the narrow sense P and global productivity P* (Baayen 1989, 1992, 1993), as well as frequency, which is calculated in terms of type and token (Bauer 2005). Two conclusions are reached. Firstly, the indexes of producti- vity P and global productivity P* of -ettan are higher than the ones of -laecan. After a comparison with other Old English affixes, it turns out that -ettan is relatively produc- tive while the suffix -laecan is practically unproductive. Secondly, the indexes of pro- ductivity and the frequencies of type and token of these affixes coincide and are in accordance with Bauer's (2004) generalization that a low type frequency matched by a high token frequency implies low productivity.
The Slovene language is often presented as a national element. Even in the 19th century, which saw the Spring of Nations and the United Slovenia project, the Slovene language was a constitutive element of the Slovene nation. In the meantime, the Slovene language was positioning itself as an all-Slovene language, trying to be supra-regional. By the end of the 19th and early 20th centuries, the Slovene written language had stabilized, while at the same time the spoken language had only begun to assert itself. During this time, the prevailing principle was to "speak the way the language is written." In the mid-20th century, the theoretical idea of a literary language that is based on the central Slovene-speech (i.e. the speech of Ljubljana) came to dominate. In the third millennium, the question is whether a regionally-defined speech can be used as the basis for a Standard language. Another central question is what this "suitable" regionally-conditioned speech would be like. The principle of how important, decision-wise, the centre of a nation is, when it comes to questions of linguistic norms, may seem very attractive and, to a certain extent, logical. However, even examples of historically and linguistically comparable languages do not support the theory of creating the norm for the Standard Slovene language, based on the contemporary speech of Ljubljana, as claimed by Toporišič in Slovenska slovnica and, later, in Slovenski pravopis. Within Slovenia, the Standard Slovene language is tied to written language, which has proven, in the past, to be a suitable way of setting the norm. Regressing back to the principles of standardising a language, based on regional variants, would be unproductive, would introduce needless discord, and would cause problems with everyday, public communication. Contemporary research of actual speech, a portion of which is also presented within this article, confirms the all-Slovene and regionally-independent character of the Slovene Standard language.
Starting from the definition of treebanks and considering that treebanks are theory dependent, we propose an annotation scheme for Romanian using several approaches ranging from phrase structure to dependency grammars and property grammars. The annotation has its starting point in a generative grammar study of the Romanian AP and validates the data of the linguistic study using an annotation scheme consisting of a constraint based approach.
The degree of translation adequacy and full-value depends on its compliance with the existing general linguistic norms. Vocabulary potential of a translator is determined by the proficiency of language to translate into. Key moments in course of transferring means of another language text are those three main features: context, word-collocations, the knowledge of ethnic specifications. Meanings of words and sentences and even whole abstracts are not autonomous, and depend on the general distributions and surroundings.
In the past few years, much attention has been paid on extending phrase-based statistical machine translation with syntactic structures. In this paper we introduce a novel syntax encapsulated phrase(SEP) model, in which treebank tag sequences are employed to decorate the bilingual phrase pairs. We use tag sequences, instead of phrase pairs, to train the lexicalized reordering model. Since the number of treebank tags is much smaller than the number of words, the tag sequence based reordering model is smaller and more accurate than the phrase based reordering model. Experiments were carried out on four types of models: the phrase model, the hierarchical phrase model, the POS tag encapsulated phrase(PTEP) model and the syntactic tag encapsulated phrase(STEP) model. The STEP model obtained higher BLEU-4 score than other models on NIST 2005 MT task.
Learning vocabulary and understanding texts present difficulty for language learners due to, among other things, the high degree of lexical ambiguity. By developing an intelligent tutoring system, this dissertation examines whether automatically providing enriched sense-specific information is effective for vocabulary learning and reading comprehension of second language learners. The system developed in this study contributes to an extended understanding of how NLP techniques can be applied more effectively in an educational environment. The system allows learners to upload texts and click on any content word in order to obtain sense-appropriate lexical information for unfamiliar or unknown words during reading. The system consists of three components: (1) the system manager controls the interaction among each learner, the NLP server, and the lexical database; (2) the NLP server converts a raw input text to a linguistically-analyzed text; (3) the lexical database is used to provide a sense-appropriate definition and example sentences of a word to the learner. To obtain the sense-appropriate information, the system first performs word sense disambiguation (WSD) on the input text. Pointing to appropriate examples tuned for language learners, however, is complicated by the fact that the database of examples is from one repository (COBUILD), while automatic WSD systems generally rely on senses from another (WordNet). The lexical database, then, is indexed by WordNet senses, each of which points to an appropriate corresponding COBUILD sense. The fact that every sense inventory has its own standards of sense distinction poses a serious problem in integrating these inventories into one. To redirect an input WordNet sense to a corresponding COBUILD sense, thus, a word sense alignment algorithm was developed, following a heuristic of favoring flatter alignment structures. With this system, an empirical study was conducted with 60 intermediate learners of English as a second language to examine whether this system can lead learners to improve their vocabulary acquisition and reading comprehension. The findings show that learners demonstrated higher performance when receiving sense-specific information. Furthermore, the qualitative examination of the effect of automatic system errors show that, although learners showed learning regardless of the appropriateness of lexical information, they still showed relatively greater learning when given appropriate lexical information.
We propose HamleDT – HArmonized Multi-LanguagE Dependency Treebank. HamleDT is a compilation of existing dependency treebanks (or dependency conversions of other treebanks), transformed so that they all conform to the same annotation style. While the license terms prevent us from directly redistributing the corpora, most of them are easily acquirable for research purposes. What we provide instead is the software that normalizes tree structures in the data obtained by the user from their original providers. Keywords:dependency treebank, annotation scheme, harmonization 1.
Here we describe work on learning the subcategories of verbs in a morphologically rich language using only minimal linguistic resources. Our goal is to learn verb subcategorizations for Quechua, an under-resourced morphologically rich language, from an unannotated corpus. We compare results from applying this approach to an unannotated Arabic corpus with those achieved by processing the same text in treebank form. The original plan was to use only a morphological analyzer and an unannotated corpus, but experiments suggest that this approach by itself will not be effective for learning the combinatorial potential of Arabic verbs in general. The lower bound on resources for acquiring this information is somewhat higher, apparently requiring a a part-of-speech tagger and chunker for most languages, and a morphological disambiguater for Arabic.
The study of the Tip of the Tongue phenomenon (TOT) provides valuable clues and insights concerning the organisation of the mental lexicon (meaning, number of syllables, relation with other words, etc.). This paper describes a tool based on psycho-linguistic observations concerning the TOT phenomenon. We've built it to enable a speaker/writer to find the word he is looking for, word he may know, but which he is unable to access in time. We try to simulate the TOT phenomenon by creating a situation where the system knows the target word, yet is unable to access it. In order to find the target word we make use of the paradigmatic and syntagmatic associations stored in the linguistic databases. Our experiment allows the following conclusion: a tool like SVETLAN, capable to structure (automatically) a dictionary by domains can be used sucessfully to help the speaker/writer to find the word he is looking for, if it is combined with a database rich in terms of paradigmatic links like EuroWordNet.
在句法分析中,已有研究工作表明,词汇依存信息对短语结构句法分析是有帮助的,但是已有的研究工作都仅局限于使用一阶的词汇依存信息.提出了一种使用高阶词汇依存信息对短语结构树进行重排序的模型,该模型首先为输入句子生成有约束的搜索空间(例如,N-best 句法分析树列表或者句法分析森林),然后在约束空间内获取高阶词汇依存特征,并利用这些特征对短语结构候选树进行重排序,最终选择出最优短语结构分析树.在宾州中文树库上的实验结果表明,该模型的最高 F1 值达到了 85.74%,超过了目前在宾州中文树库上的最好结果.另外,在短语结构分析树的基础上生成的依存结构树的准确率也有了大幅提升.;The existing works on parsing show that lexical dependencies are helpful for phrase tree parsing.However, only first-order lexical dependencies have been employed and investigated in previous research. Thispaper proposes a novel method for employing higher-order lexical dependencies for phrase tree evaluation. Themethod is based on a parse reranking framework, which provides a constrained search space (via N-best lists orparse forests) and enables the parser to employ relatively complicated lexical dependency features. The models areevaluated on the UPenn Chinese Treebank. The highest F1 score reaches 85.74% and has outperformed allpreviously reported state-of-the-art systems. The dependency accuracy of phrase trees generated by the parser hasbeen significantly improved as well.
It is my pleasure to introduce this thematic issue dedicated to the lexicography of Japanese as a second or foreign language, the first thematic issue in Acta Linguistica Asiatica since its inception.Japanese has an outstandingly long and rich lexicographical tradition, but there have been relatively few dictionaries of Japanese targeted at learners of Japanese as a foreign or second language until the end of the twentieth century. With the growth of Japanese language teaching and learning around the world, the rapid development of very large scale linguistic resources and language processing technologies for Japanese, a new generation of aggregated, collectively developed or crowd-sourced resources evolving in the context of the social web, a shift from static paper to constantly developing electronic resources, the spread of internet access on hand-held devices, and new approaches to the use of language reference resources stemming from these developments, dictionaries and other reference resources for learners, teachers and users of Japanese as a foreign/second language are being developed and used in new ways in different user communities. However, information about such developments often does not reach researchers, lexicographers, dictionary users and language teachers in other user communities or research spheres. This special issues wishes to contribute to the spread of such information by presenting some recent developments in this growing field.Having received a very lively response to our call for papers, not all papers selected for publishing could fit into this issue, and part of them will be included in the December issue of ALA, which is also going to be dedicated to Japanese lexicography.The first round of papers included in this issue presents a varied cross-section of current JFL lexicographical work and research. All papers in this issue point out the relative scarcity of appropriate reference works for learners of Japanese as a foreign language, especially when compared to lexicographical resources for Japanese native speakers, and each of the endeavours presented here confronts this lack with its own original approach. Reflecting the paradigm shift in Japanese language research, where corpus research is again playing a central role, most papers presented here take advantage of the bounty of newly available corpora and web data, most prominent among which is the Balanced Corpus of Contemporary Written Japanese developed by the National Institute for Japanese Language and Linguistics in Tokyo, and which is used by Mogi, Pardeshi et al. and Sunakawa et al. in their lexicographical research and projects, while Blin taps data for his research from the web, another increasingly important linguistic resource.The first two papers offer two perspectives on existing Japanese dictionaries. Tom Gally in his paper Kokugo Dictionaries as Tools for Learners: Problems and Potential points out the drawbacks of currently available Japanese dictionaries from the perspective of learners of Japanese as a foreign language, but at the same time offers a very detailed and convincing explanation of the merits of monolingual Japanese dictionaries for native speakers (kokugo dictionaries), such as their comprehensiveness, detailedness and quantity of contextual information, when compared to bilingual dictionaries, which make them a potentially useful resource even for an audience they are not targeting - foreign language learners. His detailed explanation of possible uses and potential hurdles and pitfalls learners may encounter in using them, is not only accurate and informative, but also of immediate practical value for language teachers and lexicographers.Toshinobu Mogi, in his paper Towards the Lexicographic Description of the Grammatical Behaviour of Japanese Loanwords: A Case Study, investigates the lexicographic description of loanwords in Japanese reference works and notes how information offered by currently available dictionaries, especially regarding the grammatical aspects of loanword use, is not sufficient for learners of Japanese as a foreign language. After pointing our the deficiencies of current dictionary descriptions and noting how dictionaries sense divisions do not reflect the frequency of different senses in actual use, as reflected in a large-scale representative general corpus of Japanese, he uses a fascinatingly detailed analysis of the behaviour of a Japanese loanword verb to describe a corpus-based method of lexical description, based on the correspondence between usage forms and senses, which could be used for the compilation of Japanese learners' dictionaries meant for the reception and production of Japanese.The second part of this special issue is composed of four reports on particular aspects of ongoing lexicographical work targeted at learners of Japanese as a foreign language.Prashant Pardeshi, Shingo Imai, Kazuyuki Kiryu, Sangmok Lee, Shiro Akasegawa and Yasunari Imamura in their paper Compilation of Japanese Basic Verb Usage Handbook for JFL Learners: A Project Report, after pointing out - as other authors in this issue - the lack of a detailed and pedagogically sound lexicographical description of Japanese basic vocabulary for foreign learners, propose a corpus-based on-line system which incorporates insights from cognitive grammar, contrastive studies and second language acquisition research to solve this problem. They present their current implementation of such a system, which includes audio-visual material and translations into Chinese, Korean and Marathi. The system also uses natural language processing techniques to support lexicographers who need to process daunting amounts of corpus data in order to produce detailed lexical descriptions based on actual use.The next article by Marcella Maria Mariotti and Alessandro Mantelli, ITADICT Project and Japanese Language Learning, focus on the learner's perspective. They present a collaborative project in which Italian learners of Japanese compiled an on-line Japanese-Italian dictionary using a purposely developed on-line dictionary editing system, under the supervision of a small group of teachers. One practical and obvious outcome of the project is a Japanese-Italian freely accessible lexical database, but the authors also highlight the pedagogical value of such an approach, which stimulates students' motivation for learning, hones their ICT skills, makes them more aware of the structure and usability of existing lexicographic and language learning resources, and helps them learn to cooperate on a shared task and exchange peer support.The third project report by Raoul Blin, Automatic Addition of Genre Information in a Japanese Dictionary, focuses on the labelling of lexical genre, an aspect of word usage which is not satisfactorily presented in current Japanese dictionaries, despite its importance for foreign language learners when using dictionaries for production tasks. The article describes a procedure for automatic labelling of genre by means of a statistical analysis of internet-derived genre-specific corpora. The automatisation of the process simplifies its later reiteration, thus making it possible to observe lexical genre development over time.The final paper in this issue is a report on The Construction of a Database to Support the Compilation of Japanese Learners’ Dictionaries, by Yuriko Sunakawa, Jae-ho Lee and Mari Takahara. Motivated by the lack of Japanese bilingual learners' dictionaries for speakers of most languages in the world, the authors engaged in the development of a database of detailed corpus-based descriptions of the vocabulary needed by learners of Japanese from beginning to advanced level. By freely offering online the basic data needed for bilingual dictionary compilation, they are building the basis from which editors in under-resourced language areas will be able to compile richer and more up-to-date contents even with limited human and financial resources. This project is certainly going to greatly contribute to the solution of existing problems in Japanese learners' lexicography.
We introduce a substantial update of the Prague Czech-English Dependency Treebank, a parallel corpus manually annotated at the deep syntactic layer of linguistic representation. The English part consists of the Wall Street Journal (WSJ) section of the Penn Treebank. The Czech part was translated from the English source sentence by sentence. This paper gives a high level overview of the underlying linguistic theory (the so-called tectogrammatical annotation) with some details of the most important features like valency annotation, ellipsis reconstruction or coreference.
This paper reports the annotation of a Brazilian Portuguese Treebank with semantic role labels following Propbank guidelines. A different language and a different parser output impact the task and require some decisions on how to annotate the corpus. Therefore, a new annotation guide – called Propbank-Br- has been generated to deal with specific language phenomena and parser problems. In this phase of the project, the corpus was annotated by a unique linguist. The annotation task reported here is inserted in a larger projet for the Brazilian Portuguese language. This project aims to build Brazilian verbs frames files and a broader and distributed annotation of semantic role labels in Brazilian Portuguese, allowing inter-annotator agreement measures. The corpus, available in web, is already being used to build a semantic tagger for Portuguese language.
As interest grows in the use of linguistically annotated corpora in research and teaching of foreign languages and literature, treebanks of various historical texts have been developed. We introduce the first large-scale dependency treebank for Classical Chinese literature. Derived from the Stanford dependency types, it consists of over 32K characters drawn from a collection of poems written in the 8 th century CE. We report on the design of new dependency relations, discuss aspects of the annotation process and evaluation, and illustrate its use in a study of parallelism in Classical Chinese poetry. 1
The paper introduces an ongoing project for the development of a parallel treebank for Italian, English and French, i.e. Parallel–TUT, or simply ParTUT. For the development of this resource, both the dependency and constituency-based formats of the Italian Turin University Treebank (TUT) have been applied to a preliminary dataset, which includes the whole text of the Universal Declaration of Human Rights, sentences from the JRC-Acquis Multilingual Parallel Corpus and the Creative Commons licence. The focus of the project is mainly on the quality of the annotation and the investigation of some issues related to the alignment of data that can be allowed by the TUT formats, also taking into account the availability of conversion tools for display data in standard ways, such as Tiger–XML and CoNLL formats. It is, in fact, our belief that increasing the portability of our treebank could give us the opportunity to access resources and tools provided by other research groups, especially at this stage of the project, where no particular tool – compatible with the TUT format – is available in order to tackle the alignment problems.
The paper describes the methodology which is currently being defined for the construction of a “Merged Italian Dependency Treebank” (MIDT) starting from already existing resources. In particular, it reports the results of a case study carried out on two available dependency treebanks, i.e. TUT and ISST–TANL. The issues raised during the comparison of the annotation schemes underlying the two treebanks are discussed and investigated with a particular emphasis on the definition of a set of linguistic categories to be used as a “bridge” between the specific schemes. As an encoding format, the CoNLL de facto standard is used.
Logical Forms are an exceptionally important linguistic representation for highly demanding semantically related tasks like Question / Answering and Text Understanding, but their automatic production at runtime is higly error-prone. The use of a tool like XWNet and other similar resources would be beneficial for all the NLP community, but not only. The problem is: Logical Forms are useful as long as they are consistent, otherwise they would be useless if not harmful. Like any other resource that aims at providing a meaning representation, LFs require a big effort in manual checking order to reduce the number of errors to the minimum acceptable – less than 1 %- from any digital resource. As will be shown in detail in the paper, the available resources – XWNet, WN30-lfs, ILF-suffer from lack of a careful manual checking phase, and the number of errors is too high to make the resource usable as is. We classified mistakes by their syntactic or semantic type in order to facilitate a revision of the resource that we intend to do using regular expressions. We also commented extensively on semantic issues and on the best way to represent them in Logical Forms. 1.
In this paper we present and justify methodological principles and syntactic criteria to design an annotation scheme for a Persian Treebank. The advantages of the proposed scheme for annotation of the Persian Treebank will be discussed. At the same time, we present the way that different types of linguistic knowledge (morphological, syntactic and semantic) are encoded in the structures of the schema. We will show how this scheme can account for many of the syntactic constructions that appear to be unique to the Persian language.
Unknown words, or out of vocabulary words (OOV), cause a significant problem to morphological analysers, syntactic parses, MT systems and other NLP applications. Unknown words make up 29 % of the word types in in a large Arabic corpus used in this study. With today's corpus sizes exceeding 10 9 words, it becomes impossible to manually check corpora for new words to be included in a lexicon. We develop a finite-state morphological guesser and integrate it with a machine-learning-based pre-annotation tool in a pipeline architecture for extracting unknown words, lemmatizing them, and giving them a priority weight for inclusion in a lexical database. The processing is performed on a corpus of contemporary Arabic of 1,089,111,204 words. Our method is tested on a manually-annotated gold standard and yields encouraging results despite the complexity of the task. Our work shows the usability of a highly
State-of-the-art dependency representations such as the Stanford Typed Dependencies may represent the grammatical relations in a sentence as directed, possibly cyclic graphs. Querying a syntactically annotated corpus for grammatical structures that are represented as graphs requires graph matching, which is a non-trivial task. In this paper, we present an algorithm for graph matching that is tailored to the properties of large, syntactically annotated corpora. The implementation of the algorithm is built on top of the popular IMS Open Corpus Workbench, allowing corpus linguists to re-use existing infrastructure. An evaluation of the resulting software, CWB-treebank, shows that its performance in real world applications, such as a web query interface, compares favourably to implementations that rely on a relational database or a dedicated graph database while at the same time offering a greater expres-sive power for queries. An intuitive graphical interface for building the query graphs is available via the Treebank.info project.
This paper presents the work of the Hong Kong Polytechnic University (PolyUCOMP) team which has participated in the Semantic Textual Similarity task of SemEval-2012. The PolyUCOMP system combines semantic vectors with skip bigrams to determine sentence similarity. The semantic vector is used to compute similarities between sentence pairs using the lexical database WordNet and the Wikipedia corpus. The use of skip bigram is to introduce the order of words in measuring sentence similarity. 1
This paper describes the support for mouth activity annotation provided by the iLex annotation workbench on a holistic level connected to the lexical database, on a feature level, as well as in the context of semi-automatic annotation.
Query expansion is a crucial step in recall-oriented domains such as Patent Searching. Currently, automatic query expansion in patent search is mostly based on statistical measures. Additional query terms are extracted from the query documents based on entropy measures. To automate query expansion in patent searching, we acquire lexical knowledge from Query Logs of USPTO Patent Examiners. Results show good performance in query expansion and patent searching using the lexical database. This will help improving (semi-) automated query expansion in patent searching.
Treebanking a large corpus of relatively structured speech transcribed from various Arabic Broadcast News (BN) sources has allowed us to begin to address the many challenges of annotating and parsing a speech corpus in Arabic. The now completed Arabic Treebank BN corpus consists of 432,976 source tokens (517,080 tree tokens) in 120 files of manually transcribed news broadcasts. Because news broadcasts are predominantly scripted, most of the transcribed speech is in Modern Standard Arabic (MSA). As such, the lexical and syntactic structures are very similar to the MSA in written newswire data. However, because this is spoken news, cross-linguistic speech effects such as restarts, fillers, hesitations, and repetitions are common. There is also a certain amount of dialect data present in the BN corpus, from on-the-street interviews and similar informal contexts. In this paper, we describe the finished corpus and focus on some of the necessary additions to our annotation guidelines, along with some of the technical challenges of a treebanked speech corpus and an initial parsing evaluation for this data. This corpus will be available to the community in 2012 as an LDC publication.
Continuous self-reported emotion expressed by four pieces of music were collected on a two-dimensional (valence and arousal) emotion space in a repeated measures (test-retest conditions) design. Initial orientation time (IOT), test-retest reliability and afterglow were examined. Median IOT was 8 seconds. Valence ratings took up to 25 (median 4), and for arousal up to 35 (median 12) seconds. Slower tempi seemed to require longer IOT. Test-retest reliability examined correlation coefficients, and compared periods of sample-by-sample good agreement in response between Test and Retest condition. About 80% of responses were reliable in both the Test and Retest conditions regardless of response dimension. Pearson correlations demonstrated better test-retest reliability for arousal responses than for valence. Retest condition ratings were within 8% of Test condition rating within participant. Average standard deviations for ratings collapsed across dimension, stimulus and conditions was 12.2% of the ratings scale range. Afterglow effects – large outliers in spread of scores just after the end of a piece – were identified. The reliability of continuous emotional response is therefore considered to be quite good, but caution must be taken as to how to deal with the opening and ending of continuous emotional response data.
The lack of annotated corpora brings limitations in research of discourse classification for many languages. In this paper, we present the first effort towards recognizing ambiguities of discourse connectives, which is fundamental to discourse classification for resource-poor language such as Chinese. A language independent framework is proposed utilizing bilingual dictionaries, Penn Discourse Treebank and parallel data between English and Chinese. We start from translating the English connectives to Chinese using a bi-lingual dictionary. Then, the ambiguities in terms of senses a connective may signal are estimated based on the ambiguities of English connectives and word alignment information. Finally, the ambiguity between discourse usage and non-discourse usage were disambiguated using the co-training algorithm. Experimental results showed the proposed method not only built a high quality connective lexicon for Chinese but also achieved a high performance in recognizing the ambiguities. We also present a discourse corpus for Chinese which will soon become the first Chinese discourse corpus publicly available.
While it is possible to build a formal grammar manually from scratch or, going to another extreme, to derive it automatically from a treebank, the development of the LFG grammar of Polish presented in this paper is different from both of these methods as it relies on extensive reuse of existing language resources for Polish. LFG grammars minimally provide two levels of representation: constituent structure (c-structure) produced by context-free phrase structure rules and functional structure (f-structure) created by functional descrip-tions. The c-structure was based on a DCG grammar of Polish, while the f-structure level was mainly inspired by the available HPSG analyses of Polish. The morphosyntactic information needed to create a lexicon may be taken from one of the following resources: a morphological analyser, a treebank or a corpus. Valence information from the dictionary which accompanies the DCG grammar was con-verted so that subcategorisation is stated in terms of grammatical functions rather than categories; additionally, missing valence frames may be extracted from the treebank. The obtained grammar is evaluated using constructed testsuites (half of which were provided by previous grammars) and the treebank.
Après un bref rĂŠsumĂŠ de la thĂŠorie de Topic-Focus Articulation (TFA), la prĂŠsente ĂŠtude dĂŠmontre, à l'aide de plusieurs exemples illustrant l'annotation de principaux traits de TFA sur un large corpus (the Prague Dependency Treebank), que l'annotation du corpus apporte une valeur ajoutĂŠe au corpus, si deux conditions sont rĂŠunies: (i) le schĂŠma de l'annotation est basĂŠ sur une thĂŠorie linguistique solide, (ii) le procĂŠdĂŠ d'annotation est ĂŠtabli avec soin (c'est-à-dire de façon systĂŠmatique et cohĂŠrente). Une telle annotation est importante non seulement pour la structure de surface de la phrase mais encore davantage pour la structure phrastique sous-jacente, car elle est susceptible de mettre en ĂŠvidence les phĂŠnomènes cachĂŠs au niveau de la structure de surface, mais incontournables lors de la reprĂŠsentation du sens et du fonctionnement de la phrase.
Cet article constitue une version réduite de l'article "The French Social Media Bank: a Treebank of Noisy User Generated Content" (mêmes auteurs)
International audience
The study presented in this article is dedicated to a syntactic parser for Romanian. The central goal of the presented technique is to learn a model which is able to discriminate between probability for a word to be head of another word in a dependency structure corresponding to a sentence in the considered language. The model described in this paper was trained on a dependency treebank linguistic resource and is intended to be used in order to develop a dependency syntactic parser.
Existing logic-based querying tools for dependency treebanks use first order logic or monadic second order logic. We introduce a very fast model checker based on hybrid logic with operators ↓, @ and A and show that it is much faster than an existing querying tool for dependency treebanks based on first order logic, and much faster than an existing general purpose hybrid logic model checker. The querying tool is made publicly available.
In this article, we first present the overall structure of the Pralex lexical database, the work with the data entry form and Praled’s functions. We then focus on the general principles of the elementary processing of database entries, which are subsequently specified according to the individual word classes and entry types. The article ends with a specific example of the processing of an entry in the Pralex lexical database.
Sleep enhances memories, particularly emotional memories. As such, it has been suggested that sleep deprivation may reduce posttraumatic stress disorder. This presumes that emotional memory consolidation is paralleled by a reduction in emotional reactivity, an association that has not yet been examined. In the present experiment, we used an incidental memory task in humans and obtained valence and arousal ratings during two sessions separated either by 12 h of daytime wake or 12 h including overnight sleep. Recognition accuracy was greater following sleep relative to wake for both negative and neutral pictures. While emotional reactivity to negative pictures was greatly reduced over wake, the negative emotional response was relatively preserved over sleep. Moreover, protection of emotional reactivity was associated with greater time in REM sleep. Recognition accuracy, however, was not associated with REM. Thus, we provide the first evidence that sleep enhances emotional memory while preserving emotional reactivity.
The goal of the presented parallel phrase extraction algorithm is to provide rich and robust set of translation syntactic patterns. To make this approach feasible, we consider the phrase-to-phrase alignments of a bilingual treebank annotated with syntactic constituents. For the intended purpose, the extracted phrasal nodes are encoded by the syntactical information of their components, highlighting some special constructs such as the functional words.
The recent construction of large linguistic treebanks for spoken and written Dutch (e.g. CGN, LASSY, Alpino) has created new and exciting opportunities for the empirical investigation of Dutch syntax and semantics. However, the exploitation of those treebanks requires knowledge of specific data structures and query languages such as XPath. Linguists who are unfamiliar with formal languages are often reluctant towards learning such a language. In order to make treebank querying more attractive for non-technical users we developed GrETEL (Greedy Extraction of Trees for Empirical Linguistics), a query engine in which linguists can use natural language examples as a starting point for searching the Lassy treebank without knowledge about tree representations nor formal query languages. By allowing linguists to search for similar constructions as the example they provide, we hope to bridge the gap between traditional and computational linguistics. Two case studies are conducted to provide a concrete demonstration of the tool. The architecture of the tool is optimised for searching the LASSY treebank, but the approach can be adapted to other treebank lay-outs.
Mood is an important aspect of music and knowledge of mood can be used as a basic feature in music recommender and retrieval systems. A listening experiment was carried out establishing ratings for various moods and a number of attributes, e.g., valence and arousal. The analysis of these data covers the issues of the number of basic dimensions in music mood, their relation to valence and arousal, the distribution of moods in the valence-arousal plane, distinctiveness of the labels, and appropriate (number of) labels for full coverage of the plane. It is also shown that subject-averaged valence and arousal ratings can be predicted from music features by a linear model.
We also show ensemble dependency parsing and self training approaches applicable to under-resourced languages using our manually annotated dependency structures. We show that for an under-resourced language, the use of tuning data for a meta classifier is more effective than using it as additional training data for individual parsers. This meta-classifier creates an ensemble dependency parser and increases the dependency accuracy by 4.92% on average and 1.99% over the best individual models on average. As the data sizes grow for the the under-resourced language a meta classifier can easily adapt. To the best of our knowledge this is the first full implementation of a dependency parser for Indonesian. Using self-training in combination with our Ensemble SVM Parser we show additional improvement. Using this parsing model we plan on expanding the size of the corpus by using a semi-supervised approach by applying the parser and correcting the errors, reducing the amount of annotation time needed.
Language resources are essential for linguistic research and the development of NLP applications. Low-density languages, such as Irish, therefore lack significant research in this area. This paper describes the early stages in the development of new language resources for Irish – namely the first Irish dependency treebank and the first Irish statistical dependency parser. We present the methodology behind building our new treebank and the steps we take to leverage upon the few existing resources. We discuss language-specific choices made when defining our dependency labelling scheme, and describe interesting Irish language characteristics such as prepositional attachment, copula and clefting. We manually develop a small treebank of 300 sentences based on an existing POS-tagged corpus and report an inter-annotator agreement of 0.7902. We train MaltParser to achieve preliminary parsing results for Irish and describe a bootstrapping approach for further stages of development.
This paper describes a method to convert existing treebanks with syntactic information into banks of meaning representations. The central component is a system of evaluation for a small formal language with respect to an information state. Inputs to the evaluation system are formal language expressions obtained from the conversion of parsed representations conforming to (Penn Treebank Project) guidelines. Outputs from the evaluation system are Davidsonian (higher-order) predicate logic meaning representations. Having a system of evaluation as the basis for generating meaning representations makes possible accepting input with minimal conversion from existing treebanks and from the tools used to construct treebanks. Results of having built corresponding banks of meaning representations from available treebanks are discussed.
OBJECTIVE: To prospectively compare an indocyanine green (ICG)-enhanced optical imaging system with contrast-enhanced magnetic resonance imaging (MRI) for the detection of synovitis in the hands of patients with rheumatologic disorders. METHODS: Forty-five patients (30 women [67%], mean ± SD age 52.6 ± 13.4 years) in whom there was a clinical suspicion of an inflammatory arthropathy were examined with a commercially available device for ICG-enhanced optical imaging as well as by contrast-enhanced 3T MRI as the standard of reference. Three independent readers graded the degree of synovitis in the carpal, metacarpophalangeal, proximal interphalangeal, and distal interphalangeal joints of both hands (1,350 joints), using a 4-point ordinate scale (0 = no synovitis, 1 = mild, 2 = moderate, 3 = severe). Statistical analyses were performed using a logistic generalized estimating equation approach. Agreement of optical imaging ratings made by the different readers was estimated with a weighted kappa coefficient. RESULTS: When MRI was used as the standard of reference, optical imaging showed a sensitivity of 39.6% (95% confidence interval [95% CI] 31.1-48.7%), a specificity of 85.2% (95% CI 79.5-89.5%), and accuracy of 67.0% (95% CI 61.4-72.1%) for the detection of synovitis in patients with arthritis. Diagnostic accuracy was especially limited in the setting of mild synovitis, while it was substantially better in patients with severely inflamed joints. Moderate interreader and intrareader agreement was observed. CONCLUSION: The evaluated ICG-enhanced optical imaging system showed limitations for the detection of inflamed joints of the hand in comparison with MRI.
Quantitative linguistics (QL) is a discipline of linguistics, that, using real texts, studies languages with quantitative mathematical approaches, aiming to precisely describe and explain, with a system of mathematical laws, the operation and development of language systems. Later in this review, we will address the relationship between QL and computational linguistics. Quantitative Syntax Analysis is a recent work on QL by Reinhard Köhler that not only provides a comprehensive introduction to the work of QL on the syntactic level, but also sketches the theoretical grounds, the research paradigm, and the ultimate goals of quantitative linguistics in general.In the first chapter, Köhler points to the vital role of syntax in language: Syntax enables language users to code structures instead of ideas as wholes. A text embodies a complex cognitive formation and meets several basic requirements in human communication, which implies that language is not autonomous, but a dynamic communicative system used by human beings. Hence, the ultimate understanding and explanation of syntax (and the whole language system) depends on usage-based investigation of the cognitive basis and the functional requirements of language, which is somewhat neglected in many mainstream syntactic studies. Even in those cases where explanatory power is acknowledged as the ultimate goal of linguistic investigation, the necessary knowledge is still required as to what a scientific linguistic theory is and how such a theory may be built. So far, it is rare for quantitative means to be used in syntactic study. One reason is that many syntacticians are too addicted to the enshrined traditional paradigms that have been proven to be somewhat inadequate when it comes to processing real texts. And that is why in computational linguistics, “devout executors of the belief in strictly formal methods as opposed to statistical ones do not have any chance to succeed” (page 4).The second chapter, entitled “The Quantitative Analysis of Language and Text,” begins with an explanation of the difference between quantitative linguistics and the formal branches of linguistics that have once been widely used in computational linguistics: QL is concerned with the quantitative properties important for understanding the development and the operation of linguistic systems, whereas the formal branches of linguistics use only qualitative mathematical means and formal logics to model structural properties of language, overlooking, in most cases, the aspects of systems that exceed structure, viz., functions, dynamics, and processes. Köhler points out that the successes of modern natural sciences (the exact, testable statements, the precise predictions, and the copious applications) all derive from their instruments and their advanced models. This implies that these instruments and models, for which the quantitative parts of mathematics (probability theory and statistics, function theory, differential equations) are indispensable ingredients, are worth integrating into linguistics, which is the aim of QL.Chapter 3, entitled “Empirical Analysis and Mathematical Modeling,” reviews the important works of quantitative syntactic analysis. In the first section, Köhler gives a long list of the important syntactic units and properties defined within the frameworks of both phrase structure syntax and dependency syntax; this reflects the fact that researchers in both fields have been engaged in some fruitful quantitative studies. In Section 3.2, he defines quantitation of syntactic concepts as counting the objects under study, because syntactic analysis investigates only discrete objects. Section 3.4 is a detailed review of the important works on various syntactic phenomena within the frameworks of both phrase structure syntax and dependency syntax, including sentence length, probabilistic grammars and probabilistic parsing, Markov chains, Frumkina’s law on the syntactic level, distribution of dependency distance, and distribution of dependency types, and so on. These quantitative models, which have been empirically corroborated with real texts, or sometimes treebanks and dictionaries (of various languages), can be linguistically, cognitively, or functionally interpreted—a rare achievement in the past statistical investigations of language. Apart from the models concerning probabilistic grammars and Markov chains, which have already been widely used in computational linguistics, there are some other works that may also have practical applications in various fields. For example, the mathematical model of sentence length, which describes the probability of neighboring length classes as a function of the probability of the first of the two given classes, may contribute to practical applications such as text classification and the measurement of text comprehensibility, and so forth. The frequency studies of word and syntactic constructions have obtained many results useful for language teaching, the construction of parsing algorithms, and the estimation of effort of (automatic) rule learning, and more. The syntactic studies on Frumkina’s law, which is concerned with the number of text blocks with x occurrences of a given syntactic element or category, may benefit certain types of computational text processing if specific constructions or categories can be differentiated and found automatically by their particular distributions. One advantage of QL is that all its findings are mathematically formulated and linguistically interpreted, which at least makes it possible to be used in constructing models necessary for computational linguistics.In Chapter 4, Köhler introduces his efforts to build a real “linguistic theory,” for he believes that “there is not yet any elaborated linguistic theory in the sense of the philosophy of science” (page 21). Building such a theory begins with “plausible hypotheses,” which may become laws when sufficiently attested and may then be further integrated into a coherent system. This is the process of setting up a scientific theory, as succinctly summarized in the title of this chapter: “Hypotheses, Laws, and Theories.”The first section of this chapter shows the first step toward a scientific linguistic theory, the process in which “plausible hypotheses” are deduced, interpreted, and empirically attested before finally becoming laws. In Section 2, Köhler introduces the foundation of his synergetic linguistics, which views language as a dynamic, self-organizing, and self-regulating system where the so-called enslaving principle and order parameters are the crucial elements. On this basis, the author builds a synergetic syntactic model in Section 4.2.7 with certain modeling principles. Within the framework of phrase-structure syntax, eight properties of syntactic constructions and four inventories are chosen to build this model, which are linked together by laws resulting from the verified hypotheses and subject to the regulation of some order parameters.Quantitative linguistics, which is an unfamiliar field of study for many linguists, depends heavily on real texts and mathematical tools. Therefore we believe it is worthwhile to briefly clarify the differences and the relations between QL and corpus linguistics, on one hand, and between QL and computational linguistics, on the other hand. In comparison with QL, corpus linguistics is in fact more of a research methodology rather than an independent linguistic discipline, reflecting a shift of focus from competence to performance, from introspection to empirical study. This makes both the common ground and the difference between corpus linguistics and QL, which aims to quantitatively and mathematically explore, on the basis of real texts and treebanks, the fundamental laws governing the structure and evolution of language, and integrate them into a systematic theory capable of explanation and prediction.Computational linguistics (CL) is an interdisciplinary field investigating the structure of natural languages from a formal, mathematical, and computational point of view. Compared with QL, CL seems to be more interested in research that can have direct applications in such fields as language understanding, language generation, machine translation, and so forth, than in the explanation of language structure, operation, and evolution. Traditionally, the mathematical models used in CL are derived from the linguistic theories via the process of formalization. Due to the limitations of the qualitative mathematical models, however, the statistical models, most of which so far fail linguistic interpretations, have recently become dominant in the field of computational linguistics. That is perhaps why Shuly Wintner (2009, page 641), in a “Last Words” article in this journal, called for “the return of linguistics to computational linguistics,” implying that the advances in the field of CL may ultimately lie in the advances in the understanding of language itself.Wintner’s appeal does reflect the present situation of computational linguistics: a discipline in which many works are heavily oriented towards engineering and weakly grounded in linguistics. The formal linguistic models seem now outshone by the purely statistical paradigms, as the result of their inadequacy in processing real-world languages. Of course, this means no renouncement of the value of the traditional formal models. But it is obvious that considerable updating and enrichment are necessary for these formal models if they are to play significant roles in the future. In this regard, QL, which has a solid linguistic foundation, may help by providing quantitative cues (which are linguistically interpretable) to improve the performance in NLP, as has been illustrated in the case of probabilistic grammar that ingeniously integrates quantitative, statistical devices into qualitative linguistic models. QL has provided some useful models for CL. And it is reasonable to believe that it will continue to do so in the future, though some of its achievements seem now not ready to be directly used in CL.But perhaps QL can do more than that. Though QL has not yet drawn much attention from computational linguistics, it is potentially a proper answer to Wintner’s call for the return of linguistics to computational linguistics. CL aims to replicate in computers the patterns of human language behavior, whereas QL endeavors to mathematically and quantitatively reveal the laws and the principles that govern human language behavior—this is a relation between theory and practice. The success of a scientific discipline is usually based on precise models that are well-grounded in profound understanding of the object of study. Computational linguistics is no exception. Two features of QL are hence noteworthy. One is that it aims to explain, within a certain linguistic framework, the operation and the evolution of language by uncovering the systematic cognitive and functional regulations that underlie human languages. The other is that it tries to mathematically model, with systematic and precise quantitative laws, these regulations and the resulting mechanism of language. In view of these two features, we believe that the success of QL will somehow and somewhat boost the studies in the field of CL and that the communication between QL and CL is and will be not only possible but also mutually beneficial. This is why we hold that it is worthwhile to recommend this book to researchers in the field of computational linguistics, a book presenting a panorama of QL in general and discoveries on the syntactic level in particular.
This paper presents the first efficient implementation of a weighted deductive CYK parser for Probabilistic Linear Context-Free Rewriting Systems (PLCFRSs). LCFRS, an extension of CFG, can describe discontinuities in a straightforward way and is therefore a natural candidate to be used for data-driven parsing. To speed up parsing, we use different context-summary estimates of parse items, some of them allowing for A* parsing. We evaluate our parser with grammars extracted from the German NeGra treebank. Our experiments show that data-driven LCFRS parsing is feasible and yields output of competitive quality.
Sophisticated models have been developed for joint word segmentation and part-of-speech tagging, with increasing accuracies reported on the Chinese Treebank data. These systems, which rely on supervised learning, typically perform worse on texts from a different domain, for which little annotation is available. We consider self-training and character clustering for domain adaptation. Both methods use only unannotated target-domain data, and are relatively straightforward to implement upon a baseline supervised system. Our results show that both methods can effectively improve target-domain performance. In addition, a combination of the two orthogonal methods leads to further improvement.