Kinds Of Corpora Research Articles

Élaboration d’une liste de vocabulaire dans un contexte FLE

In this article, we discuss the elaboration and use of vocabulary lists aimed for learners of French as a for-eign language. These lists are commonly based on corpora, which, in the ideal case, are representative, relevant and large. As for the English language, this kind of corpora has been available for a long time (e.g. the COCA and the BNC). Vocabulary lists, which are often used in learning contexts, have been based on these corpora. The situation is, however, less favorable when it comes to the French language, with fewer corpora meeting the mentioned criteria, and, thus, fewer possibilities to create vocabulary lists that are useful for learners. In this contribution, we present work that has been done in order to create a vocabulary list, Riksprovsordlistan, containing about 4,000 words and used at all Swedish universities. The discussion focuses on methodological challenges such as choice of counting unit – lemma vs. word family –, the role of frequency, thematic vocabulary, as well as characteristics of written vs. spoken corpora.

ТЕЛЕСНЫЕ ОБРАЗЫ ОЧЕЛОВЕЧЕННОГО ПРОСТРАНСТВА У КОРЕННЫХ НАРОДОВ ЮГРЫ

The problem of a person's perception of himself in the world has always been socially significant. Studies of the history of ideas about space are most significant for our vast Russia. The processes of their formation are associated with both ancient and modern times. The objective of the publication is to analyze the mental modeling of space as a kind of corpora in the traditional culture of the indigenous peoples of Yugra. To achieve it, it is necessary to resolve the following tasks: to consider the aspects of mutual evolution of the images of human corporality and the outside world among the Ob-Ugric peoples of Yugra on the basis of a preliminary reconstruction of the causes of the emergence of corporal representations in antiquity. The methodological basis of the study is an interdisciplinary approach using a number of theoretical provisions of anthropology. According to these provisions, in the processes of anthropogenesis our ancestors were able to adapt to the external environment through corporal imitation to representatives of the animal world. In turn, the naive zoomorphism of man was mentally supplemented by the anthropomorphism of the surrounding world. Ethnographic materials on the traditional culture of the peoples of Yugra testify that the properties of the human body were mentally transferred to the natural world and then to social relations. The transfer was carried out through myths. The obtained results prove the inconsistency and ambiguity of the interaction of corporal images, objects of nature and mythical characters of the Ob-Ugric folklore. We assume that the direction of further research will be related to the consideration of the specifics of the historical and cultural evolution of the images of the social corpora in the humanized space of outlying lands, among which Yugra can be attributed.

Clausal complexity of expert and student writing: a corpus-based analysis of papers in social sciences

Abstract Syntactic complexity has been extensively approached in the fields of corpus linguistics and academic discourse studies. However, works focusing on disciplinary variation in terms of linguistic complexity and comparison of professional and novice academic writing are scarce. Addressing these issues is likely to have important implications for EAP/ESP practitioners in terms of selection of target structures and learning material design. This study is a corpus analysis of the use of clausal complexity features in two social sciences, management and economics. The research is based on two kinds of corpora: expert corpora which comprise articles published in peer-reviewed journals, and learner corpora of L2 undergraduate students’ research papers. This work aims at answering two questions: Do clausal complexity features vary in the texts in management and economics? What are the differences in syntactic use between the academic texts written by professional authors and learner writing? The data showed that economists and managers tend to use the structures under consideration with significantly different frequencies. The professional writing was found to demonstrate more signs of clausal complexity than the learners’ texts.

Q-Learning for Shift-Reduce Parsing in Indonesian Tree-LSTM-Based Text Generation

Tree-LSTM algorithm accommodates tree structure processing to extract information outside the linear sequence pattern. The use of Tree-LSTM in text generation problems requires the help of an external parser at each generation iteration. Developing a good parser demands the representation of complex features and relies heavily on the grammar of the corpus. The limited corpus results in an insufficient number of vocabs for a grammar-based parser, making it less natural to link the text generation process. This research aims to solve the problem of limited corpus by proposing the use of a Reinforcement Learning algorithm in the formation of constituency trees, which link the sentence generation process given a seed phrase as the input in the Tree-LSTM model. The tree production process is modeled as a Markov’s decision process, where a set of states consists of word embedding vectors, and a set of actions of {Shift, Reduce}. The Deep Q-Network model as an approximator of the Q-Learning algorithm is trained to obtain optimal weights in representing the Q-value function. The test results on perplexity-based evaluation show that the proposed Tree-LSTM and Q-Learning combination model achieves values 9.60 and 4.60 for two kinds of corpus with 205 and 1,000 sentences, respectively, better than the Shift-All model. Human evaluation of Friedman test and posthoc analysis showed that all five respondents tended to give the same assessment for the combination model of Tree-LSTM and Q-Learning, which on average outperforms two other nongrammar models, i.e., Shift-All and Reduce-All.

The Use of Corpora in Translation Into the Second Language: A Project-Based Approach

This manuscript investigates to what extent the use of corpora could help translation trainees while translating from Arabic into English and vice versa. Forty Yemeni trainees, who were enrolled in an advanced course in Arabic-English translation during the academic year 2020, participated in the study. They participated in translation projects from which the data for this study was collected, using thinking aloud protocols and computational observation. The translation process was investigated using the translation process software Transalog, an eye-tracking software and the screen recording software Screen-O-Matic. This kind of computational observation enabled a researcher to discover the extent to which the participants were able to employ corpora in their translation projects. At the end of the study, the participants were given a questionnaire with the aim of finding out their perceptions toward the use of corpora in their translation projects, and toward the project-based training approach adopted in the study. The findings of the translation process indicated that the trainees employed various kinds of corpora in their translation projects. Results from the questionnaire showed that the trainees have very positive attitudes toward the progress in their instrumental translation sub-competence, the utilization of corpora tools, and the project-based training approach adopted in this study.

Innovatively Fused Deep Learning with Limited Noisy Data for Evaluating Translations from Poor into Rich Morphology

Evaluation of machine translation (MT) into morphologically rich languages has not been well studied despite its importance. This paper proposes a classifier, that is, a deep learning (DL) schema for MT evaluation, based on different categories of information (linguistic features, natural language processing (NLP) metrics and embeddings), by using a model for machine learning based on noisy and small datasets. The linguistic features are string based for the language pairs English (EN)–Greek (EL) and EN–Italian (IT). The paper also explores the linguistic differences that affect evaluation accuracy between different kinds of corpora. A comparative study between using a simple embedding layer (mathematically calculated) and pre-trained embeddings is conducted. Moreover, an analysis of the impact of feature selection and dimensionality reduction on classification accuracy has been conducted. Results show that using a neural network (NN) model with different input representations produces results that clearly outperform the state-of-the-art for MT evaluation for EN–EL and EN–IT, by an increase of almost 0.40 points in correlation with human judgments on pairwise MT evaluation. It is observed that the proposed algorithm achieved better results on noisy and small datasets. In addition, for a more integrated analysis of the accuracy results, a qualitative linguistic analysis has been carried out in order to address complex linguistic phenomena.

Le mot fresh en français et en tchèque – analyse contrastive des corpus dans une perspective culinaire

The word fresh in French and Czech: A contrastive analysis of corpora in a culinary perspective The main goal of this paper is to compare a sample of neological loanwords from the gastronomic sector in French and Czech. The idea of the comparative conception mentioned in the title of the article is based on the international project entitled “EmpNéo” (Neological Loanwords), which aims to compare the diffusion of neological loanwords in various languages. In this study, we would like to focus on the word fresh and its perception in Czech and French. Our aim is to find out if this word is known to the general public or, on the contrary, it works as an argotic expression, which is only known to a limited audience. To answer this question, we analyzed the different kinds of corpus and also studied in detail the occurrence of this word in the press.

Proposing a Customised Method for Extratextual Documentative Annotation on Written Text Corpus

In this paper, we have made an attempt to portray a perceivable sketch of extratextual documentative annotation which, in the present frame of text annotation, is considered as one of the indispensable processes through which we can add representational information to the texts included in a written corpus. This becomes more important when a corpus is made with a large number of texts obtained from different genres and text types. To develop a workable frame for extratextual annotation, at each stage, we have broadly classified the existing processes of corpus annotation into two broad types. Moreover, we have tried to explain different layers that are embedded with extratextual annotation of texts as well as marked out the applications which can substantially enhance the accessibility of language data from a corpus for the works of text file management, information retrieval, lexical items extraction, and language processing. The techniques that we have proposed and described in this paper are unique in the sense that these are highly useful for expanding the utility of data of a written text corpus beyond the immediate horizons of language processing to the realms of theoretical, descriptive, and applied linguistics. In this paper, we have also argued that we should try to annotate all kinds of written text corpora so far developed in different natural languages at the extratextual level in a uniform manner so that the text samples stored in corpora can be uniformly used for various works of descriptive linguistics, theoretical linguistics, language technology, and applied linguistics including grammar writing, dictionary compilation, and language teaching. The annotation scheme proposed here is applied on a sample Bangla text corpus and we have noted that the accessibility of data and information from this kind of corpus is far easier than that of an un-annotated raw corpus.

CORESPI e CORITE, due nuovi strumenti per l’analisi dell’interlingua di lingue affini

In this paper, two parallel longitudinal corpora of interlingua are presented: the CORpus del ESPanol de los Italianos (CORESPI, 474 texts, 124 648 words) and the CORpus del Italiano de los Espanoles (CORITE, 385 texts, 103 147 words). The texts have been produced by 45 couples of Spanish and Italian as foreign language learners (A1- B2). The manner in which the texts have been collected, tagged and put on the Internet using the tool CATMA are illustrated. CATMA potentialities are briefly surveyed. The usefulness of this kind of corpora for the linguistic and parallel analysis and for language teaching theory is emphasized.

Is there a Space for Post-Colonial Theory in the Socio-Psychological Research on Consequences of Colonial Past?

The focus of my commentary is two-fold. First, I discuss what appeared to me as the central theoretical focus of the article; the possibility to create a space, if at all, for integrating post-colonial theory into the broader research field of social and psychological studies of the consequences of colonial past. Second, I intend to show why, in my opinion, the methodological choices of the authors and the criteria adopted for corpus construction allowed for data that, although too thin to establishing the state of knowledge in the field of study on consequences of colonial past, is nevertheless very informative and thoughts-provoking. My conclusions suggest that this study is an innovative attempt at describing and grasping the results of a search guided by two among the more consolidated electronic datasets currently available for English-speaking scholars. However, this study may not easily understand which can be the space to integrate post-colonial theory in the field of research on consequences of colonial past. To better reach this aim, it is perhaps necessary to build another kind of corpus, open to other languages (starting from French) and focused also on other scientific products, as books or proceedings of congress. In addition, disciplinary boundaries have to be even more explored, starting from interdisciplinary studies on education and historical culture. In spite of these limitations, I am convinced that this innovative study by Tomicic and Berardi tackles issues of relevance to any serious effort towards reflecting on long-term consequences of colonial violence and opens up to valuable new research questions and methods, to be taken into serious account and further explored in future works.

Unsupervised Medical Entity Recognition and Linking in Chinese Online Medical Text.

Online medical text is full of references to medical entities (MEs), which are valuable in many applications, including medical knowledge-based (KB) construction, decision support systems, and the treatment of diseases. However, the diverse and ambiguous nature of the surface forms gives rise to a great difficulty for ME identification. Many existing solutions have focused on supervised approaches, which are often task-dependent. In other words, applying them to different kinds of corpora or identifying new entity categories requires major effort in data annotation and feature definition. In this paper, we propose unMERL, an unsupervised framework for recognizing and linking medical entities mentioned in Chinese online medical text. For ME recognition, unMERL first exploits a knowledge-driven approach to extract candidate entities from free text. Then, the categories of the candidate entities are determined using a distributed semantic-based approach. For ME linking, we propose a collaborative inference approach which takes full advantage of heterogenous entity knowledge and unstructured information in KB. Experimental results on real corpora demonstrate significant benefits compared to recent approaches with respect to both ME recognition and linking.

Construction of Jiangxi Tourism Resource E-C/C-E Parallel Corpus

This paper discusses the construction of Jiangxi Tourism Resource E-C Parallel Corpus on the basis of description of the necessity in its construction. It first describes its framework, then its construction. In construction, it includes data collection, the standard for the sorted data, data selection, data digitalization, data tagging and data aligning. With the introduction of its construction, it realizes its purposes and functions of the corpus, but also provides others with ways or means to use the corpus and to establish such kind of corpus.

POS Tagging without a Tagger: Using Aligned Corpora for Transferring Knowledge to Under-Resourced Languages

Almost all languages lack sufficient resources and tools for developing Human Language Technologies (HLT). These latters are mostly concerned by languages for which large resources and tools are available. In this paper, we will prove that under-resourced languages can benefit from these available resources and tools to develop their own HLT by taking as an example the which of the POS tagging Task that is among the most primordial Natural Language Processing tasks. Since, it assigns word tag to highlight its syntactic features by considering the corresponding contexts. The solution that we propose, in this research work, is based on the use of aligned parallel corpus as a bridge between a rich-resourced language and an under resourced language. This kind of corpus is usually available. The rich language side of this corpus is first annotated. These POS-annotations were then exploited to predict the annotation of under-resourced language side by using alignment training. After this training step, we obtain a matching table between the two languages which will be exploited to annotate an input text. The experimentation of the proposed approach is performed on a couple of languages: English as a rich language and Arabic as an under resourced language. We used the IWSLT10 training corpus, and English Treetagger. The approach was evaluated on the test corpus extracted from the IWSLT08 obtain a F-score of 89% and can be extrapolated to the other NLP tasks.

Descripció fraseològica a partir de l'anàlisi basada en un corpus paral·lel: possibilitats, límits i una proposta

Summary: The use of monolingual corpora in lexicography is now common practice. However, when creating bilingual dictionaries, the question arises to what extent parallel corpora could be sources to be taken into account. It can be assumed that this kind of corpora could provide possible contextualized lexical correspondences, if the corpora were large enough. Their use could point to problems of interlingual asymmetry, which have not always received the necessary attention in traditional lexicography. I present interlingual and context-motivated problems, which were found in the analysis of German idioms and their Catalan translation using a parallel corpus. [Keywords: parallel corpus, phraseology, interlingual asymmetries, bilingual lexicography, electronic dictionary]

PARALLEL CORPORA IN TRANSLATOR EDUCATION

Mainly four kinds of corpora have been used in the field of translator education: comparable corpora, parallel corpora, parallel learner corpora and ‘do it yourself’ or ‘disposable’ corpora. This article focuses on the use of parallel corpora in translator education. As a starting point different approaches to translator education with corpora will be described in order to figure out which one has made use of a methodological background in translation didactics to carry out their research. Secondly different parallel corpora will be presented and the strengths and limitations of different tools, such as bilingual concordancers and translation memories, will be analysed. To roundoff this article different translation alternatives to the same source text will be explored using EMEA in the context of translators’ training.

10. Job Interview Corpus: Data Transcription and Major Topics in Corpus Linguistics

The aim of this book and its accompanying audio files is to make accessible a corpus of 40 authentic job interviews conducted in English. The recordings and transcriptions of the interviews published here may be used by students, teachers and researchers alike for linguistic analyses of spoken discourse and as authentic material for language learning in the classroom. The book includes an introduction to corpus linguistics, offering insight into different kinds of corpora and discussing their main characteristics. Furthermore, major features of the discourse genre job interview are outlined and detailed information is given concerning the job interview corpus published in this book.

Mathematical Modelling of the Pattern of Occurrence of Words in Different Corpora of the Hindi Language∗

The incidence of different components of language in natural language texts are not arbitrarily organized but tend to obey particular laws which enable us to explain characteristic features of human language. The present paper is an attempt to analyse and model the pattern of occurrence of words in the Hindi language. Various kinds of corpora have been selected from different sources for the study, and the occurrence of words in these corpora has been observed for a variety of properties such as: frequencies, vocabulary measures, and pattern of initials of words relative to the subsequent matra.

Data for lexicography The central role of the corpus

This paper looks at the nature of data for lexicography and in particular on the central role that electronic corpora can play in providing it. Data has traditionally come from existing dictionaries, citations, and from the lexicographer’s own knowledge of words, through introspection. Each of these is examined and evaluated. Then the electronic corpus is considered. Different kinds of corpora are described and key design criteria are explained, in particular the size of corpus needed for lexicography as well as the issue of representativeness and sampling. The advantages and disadvantages of corpora are weighed and compared against the other types of data. While each of these has benefits, it is argued that corpora are a requirement, not an option, as data for dictionary making.

Comparative study on corpora for speech translation

This paper investigates issues in preparing corpora for developing speech-to-speech translation (S2ST). It is impractical to create a broad-coverage parallel corpus only from dialog speech. An alternative approach is to have bilingual experts write conversational-style texts in the target domain, with translations. There is, however, a risk of losing fidelity to the actual utterances. This paper focuses on balancing a tradeoff between these two kinds of corpora through the analysis of two newly developed corpora in the travel domain: a bilingual parallel corpus with 420 K utterances and a collection of in-domain dialogs using actual S2ST systems. We found that the first corpus is effective for covering utterances in the second corpus if complimented with a small number of utterances taken from monolingual dialogs. We also found that characteristics of in-domain utterances become closer to those of the first corpus when more restrictive conditions and instructions to speakers are given. These results suggest the possibility of a bootstrap-style of development of corpora and S2ST systems, where an initial S2ST system is developed with parallel texts, and is then gradually improved with in-domain utterances collected by the system as restrictions are relaxed

Constructing SSML Documents with Automatically Generated Intonation Information in a Combinatory Categorial Grammar Framework

As of now, Text-to-Speech (TTS) systems are widely used throughout the full spectrum of our activities, and various natural language processing techniques have been utilized to enhance the performance of such TTS systems. As TTS systems begin to play an important role for communication between human and machine, naturalness is considered the most crucial measure of performance for TTS systems, in addition to correctness. General statistical approaches, though widely adopted, are not appropriate for the phenomena as they assign the same intonation to the same sentence. We analyze various kinds of corpus to extract informative features for intonation generation in a Combinatory Categorial Grammar framework, and express intonation-annotated document using Speech Synthesis Markup Language for target system neutral application.