Multi-word Terms Research Articles

Čelni napad ili uzmak: hrvatsko vojno nazivlje nasuprot angloameričkomu terminološkom prodoru

With the accession of Croatia to NATO, Croatian military terminology was confronted with a vast and complex terminological pool in English, for which Croatian equivalents had to be developed. This paper aims to address and investigate both latent and conspicuous mirroring of English term-formation patterns at all linguistic levels, as well as calqueing as a mechanism for creating Croatian equivalents. The research was undertaken on a corpus-verified excerpt of terms that were subjected to terminological analysis. Prima facie observations confirmed the efforts of subject matter experts to develop contemporary Croatian military terminology using Croatian language material. However, multiple criteria revealed the sample was debatable, both from the perspective of the standard Croatian language and of terminological principles. The research demonstrated that Croatian (terminological) term formation – as the most desirable method of term formation – is significantly underrepresented and dominated by latent linguistic borrowing (calques) and multiword term units. Because multiword terms constitute the most prevalent subgroup of terms today, they were given additional consideration, especially as Anglo-American structures and patterns are most closely emulated at this level of term formation. Anglo-American formation patterns frequently take priority, typically under the pretext of their alleged precision and transparency, despite the Croatian language substantial inherent term-formation potential. To systematize military terminology and introduce stability into the conceptual system, the author deemed it beneficial to go beyond the merely descriptive level of research and render solutions for terminological gaps or the terminology requiring revision and normative preference by undertaking concrete normative interventions and providing practical paradigmatic models. Finally, the rendered solutions were designed to both catalyze future interventions in Croatian military terminology and revise the current one.

Medical terminology-based computing system: a lightweight post-processing solution for out-of-vocabulary multi-word terms

The linguistic rules of medical terminology assist in gaining acquaintance with rare/complex clinical and biomedical terms. The medical language follows a Greek and Latin-inspired nomenclature. This nomenclature aids the stakeholders in simplifying the medical terms and gaining semantic familiarity. However, natural language processing models misrepresent rare and complex biomedical words. In this study, we present MedTCS—a lightweight, post-processing module—to simplify hybridized or compound terms into regular words using medical nomenclature. MedTCS enabled the word-based embedding models to achieve 100% coverage and enabled the BiowordVec model to achieve high correlation scores (0.641 and 0.603 in UMNSRS similarity and relatedness datasets, respectively) that significantly surpass the n-gram and sub-word approaches of FastText and BERT. In the downstream task of named entity recognition (NER), MedTCS enabled the latest clinical embedding model of FastText-OA-All-300d to improve the F1-score from 0.45 to 0.80 on the BC5CDR corpus and from 0.59 to 0.81 on the NCBI-Disease corpus, respectively. Similarly, in the drug indication classification task, our model was able to increase the coverage by 9% and the F1-score by 1%. Our results indicate that incorporating a medical terminology-based module provides distinctive contextual clues to enhance vocabulary as a post-processing step on pre-trained embeddings. We demonstrate that the proposed module enables the word embedding models to generate vectors of out-of-vocabulary words effectively. We expect that our study can be a stepping stone for the use of biomedical knowledge-driven resources in NLP.

Naïve Terminological Annotation of Legal Texts in Slovak

Correct automatic terminological annotation of texts in a corpus can be sometimes a challenging task, especially for moderately or heavily inflected languages with relatively free word order. We explore the possibility of simple annotation based on sequence matching of lemmatized texts to annotate Slovak language corpus with IATE terminological entries. The accuracy of annotating legal language is very good when annotating multiword terms, while accuracy of single-word terms can be increased by applying simple filters based on word lengths and blacklisting most frequent false positives.

Multi-word terms selection for information retrieval

PurposeA number of approaches and algorithms have been proposed over the years as a basis for automatic indexing. Many of these approaches suffer from precision inefficiency at low recall. The choice of indexing units has a great impact on search system effectiveness. The authors dive beyond simple terms indexing to propose a framework for multi-word terms (MWT) filtering and indexing.Design/methodology/approachIn this paper, the authors rely on ranking MWT to filter them, keeping the most effective ones for the indexing process. The proposed model is based on filtering MWT according to their ability to capture the document topic and distinguish between different documents from the same collection. The authors rely on the hypothesis that the best MWT are those that achieve the greatest association degree. The experiments are carried out with English and French languages data sets.FindingsThe results indicate that this approach achieved precision enhancements at low recall, and it performed better than more advanced models based on terms dependencies.Originality/valueUsing and testing different association measures to select MWT that best describe the documents to enhance the precision in the first retrieved documents.

CONCEPTUAL AND TERMINOLOGICAL SYSTEM OF PEDAGOGICAL SCIENCE AS A SUBJECT SCIENTIFIC RESEARCH

The article considers the issue of nomination and interpretation of the terminological system of pedagogical science from the time of researchers’ address to this issue, ie from the 1960s to the present. The author has implemented a scientific approach, which analyzes works on both historical and contemporary aspects of pedagogical terminology in the context of pedagogical science and linguistics. The assessment of the history of development and the current state of the domestic pedagogical terminology by Ukrainian researchers-lexicologists is revealed. It was found that Ukrainian linguists in their lexicological research pay sufficient attention to the history and current state of pedagogical terminology used in pedagogical literature, including the characteristics of reference literature – pedagogical dictionaries and encyclopedias. The subject of the author’s analysis were the works of researchers-educators who analyze the terminology of pedagogical science, its positive and negative aspects in their diachronic and synchronous dimensions. In the absence of a differentiated approach in modern scientific discourse, a comparative analysis of the positions of Soviet, Russian and Ukrainian authors, who are still in a single domestic research field. It is noted that the main channel of enrichment of the conceptual and terminological apparatus of pedagogical science is modern educational practice, which does not pass the relevant expert assessment by specialists, but immediately replenishes the pedagogical terminology. It was found that the creative heritage of famous domestic teachers, their original conceptual and terminological vocabulary is beyond the scientific interests of modern Ukrainian researchers. Instead, there is an active creation of neologisms, mostly in English, new two-word or multi-word terms, especially when describing new forms and methods of teaching and education, as well as in qualifying works. The general conclusion about the need to pay more attention to research on the history and current state of the pedagogical terminology as an important component of pedagogical development. The availability of Ukrainian general pedagogical reference literature – dictionaries and encyclopedias – their terminological content also requires a separate analysis, as the development of domestic pedagogical science provides enough material to create a standardized and systematized terminology.

RECURRENT LEXIS AND PHRASEOLOGY IN ENGLISH RESTAURANT REVIEWS: A DATA-DRIVEN ANALYSIS

This paper presents a corpus linguistic analysis of recurrent vocabulary and phraseologyin written English food discourse. More specifi cally, it focuses on the use and discourse functions of keywords, key multi-word terms and lexical bundles in a specialized corpus comprising 200 professional restaurant reviews that were published in online editions of selected British and American newspapers. The results of the study indicate that the most distinctive lexical feature of the analyzed texts is the frequent mention of ingredients and the limited presence of stance devices. The most frequently mentioned aspects of the referential content also show that what is evaluated is the total experience of eating and dining at a restaurant. These fi ndings contribute to the area of English forSpecific Purposes, off ering pedagogical potential that can be exploited when developing purpose-made teaching materials for students in food-related programs who need to learn the specialized vocabulary of their target profession.

Multi-word term variation

Abstract Complex nominals (CNs) are frequently found in specialized discourse in all languages, since they are a productive method of creating terms by combining existing lexical units. In Spanish, a conceptual combination may often be rendered with a prepositional CN (PCN) or an equivalent adjectival CN (ACN), e.g., demanda de electricidad vs. demanda eléctrica [electricity demand]. Adjectives in ACNs – usually derived from nouns – are known as ‘relational adjectives’ because they encode semantic relations with other concepts. With recent exceptions, research has focused on the underlying semantic relations in CNs. In natural language processing, several works have dealt with the automatic detection of relation adjectives in Romance and Germanic languages. However, there is no discourse studies of these CNs, to our knowledge, for the goal of establishing writer recommendations. This study analyzed the co-text of equivalent PCNs and ACNs to identify factors governing the use of a certain form. EcoLexicon ES, a corpus of Spanish environmental specialized texts, was used to extract 6 relational adjectives and, subsequently, a set of 12 pairs of equivalent CNs. Their behavior in co-text was analyzed by querying EcoLexicon ES and a general language corpus with 20 expressions in CQP-syntax. Our results showed that immediate linguistic co-text determined the preference for a particular structure. Based on these findings, we provide writing guidelines to assist in the production of CNs.

The use of context in multiword-term translation

ABSTRACT The quality of a specialized translation depends to a great extent on the treatment of multiword terms (MWTs). Since MWTs are not always adequately described or even included in most terminological resources, translators and terminologists must find other ways to deal with them. One method involves the analysis of parallel texts. Traditionally, parallel texts have been a valuable resource for terminology extraction and knowledge acquisition. They can be compiled as comparable corpora (i.e. collections of texts originally written in the source and target languages). Nevertheless, even though comparable corpora can facilitate translation, many translators are often unfamiliar with corpus query techniques. This paper describes a method that identifies Spanish equivalents of English MWTs in corpora, and also provides translators and terminologists with a set of manual corpus query techniques that help to effectively translate MWTs. The comparable corpus compiled was composed of original English and Spanish texts on wind energy generation. The method proposed is based on distributional semantics and was used to identify and compare contextual elements of the concepts in both languages. The results of our study showed that this procedure simplifies and contributes to the accurate identification of MWT translation equivalents in corpora.

Detecting Multi Word Terms in patents the same way as entities

In English patent document information retrieval, Multi Word Terms (MWTs) are an important factor in determining how relevant a patent document is for a particular search query. Detecting the correct boundaries for these MWTs is no trivial task and often complicated by the special writing style of the patent domain. In this paper we describe a method for detecting MWTs in patent sentences based on a method for detecting technical entities using deep learning. On our annotated dataset of 22 patents, our method achieved an average precision of 0.75, an average recall of 0.74 and an average F1 score of 0.74. Further, we argue for the use of domain specific word embedding resources and suggest that our model mostly learns whether individual words should be included in MWTs or not.

Translating a pandemic: A corpus study of COVID-19 multi-word terminology in EU press releases

The study employs a parallel English-Polish corpus to investigate how COVID-19 multi-word terms are handled in translations of EU press releases. Translation techniques are examined at four levels of analysis: (1) term variation, (2) institutionalization, (3) domestication/foreignization, and (4) degree of transfer of information. The results are discussed in regard to the characteristics of COVID-19 terminology and its, often neological, instability, which manifests itself in high levels of terminological variation, inconsistent use of recommended institutional equivalents, and varied degrees of information transfer between SL and TL. The findings are also reviewed in light of the nature of press releases which, as an essential link in the transmission of knowledge from EU institutions to citizens, prompt the use of domesticating techniques.

Machine Learning in Terminology Extraction from Czech and English Texts

Abstract The method of automatic term recognition based on machine learning is focused primarily on the most important quantitative term attributes. It is able to successfully identify terms and non-terms (with success rate of more than 95 %) and find characteristic features of a term as a terminological unit. A single-word term can be characterized as a word with a low frequency that occurs considerably more often in specialized texts than in non-academic texts, occurs in a small number of disciplines, its distribution in the corpus is uneven as is the distance between its two instances. A multi-word term is a collocation consisting of words with low frequency and contains at least one single-word term. The method is based on quantitative features and it makes it possible to utilize the algorithms in multiple disciplines as well as to create cross-lingual applications (verified on Czech and English).

HAMLET

Abstract Automatic term extraction (ATE) is an important task within natural language processing, both separately, and as a preprocessing step for other tasks. In recent years, research has moved far beyond the traditional hybrid approach where candidate terms are extracted based on part-of-speech patterns and filtered and sorted with statistical termhood and unithood measures. While there has been an explosion of different types of features and algorithms, including machine learning methodologies, some of the fundamental problems remain unsolved, such as the ambiguous nature of the concept “term”. This has been a hurdle in the creation of data for ATE, meaning that datasets for both training and testing are scarce, and system evaluations are often limited and rarely cover multiple languages and domains. The ACTER Annotated Corpora for Term Extraction Research contain manual term annotations in four domains and three languages and have been used to investigate a supervised machine learning approach for ATE, using a binary random forest classifier with multiple types of features. The resulting system (HAMLET Hybrid Adaptable Machine Learning approach to Extract Terminology) provides detailed insights into its strengths and weaknesses. It highlights a certain unpredictability as an important drawback of machine learning methodologies, but also shows how the system appears to have learnt a robust definition of terms, producing results that are state-of-the-art, and contain few errors that are not (part of) terms in any way. Both the amount and the relevance of the training data have a substantial effect on results, and by varying the training data, it appears to be possible to adapt the system to various desired outputs, e.g., different types of terms. While certain issues remain difficult – such as the extraction of rare terms and multiword terms – this study shows how supervised machine learning is a promising methodology for ATE.

Machine Learning in Terminology Extraction from Czech and English Texts

Abstract The method of automatic term recognition based on machine learning is focused primarily on the most important quantitative term attributes. It is able to successfully identify terms and non-terms (with success rate of more than 95 %) and find characteristic features of a term as a terminological unit. A single-word term can be characterized as a word with a low frequency that occurs considerably more often in specialized texts than in non-academic texts, occurs in a small number of disciplines, its distribution in the corpus is uneven as is the distance between its two instances. A multi-word term is a collocation consisting of words with low frequency and contains at least one single-word term. The method is based on quantitative features and it makes it possible to utilize the algorithms in multiple disciplines as well as to create cross-lingual applications (verified on Czech and English).

Rule-Based Spanish Multiple Question Reformulation and their Classification using a Convolutional Neuronal Network

Question reformulation allows the creation of different forms of the same question in order to identify the best answer. However, when aspects suchas length and complexity increase, the reformulation process becomes more complicated, consequently also the recovery of the corresponding information. In this research, a method for the reformulation of multiple questions in Spanish is presented, as part of the pre-processing stage in a question-answer system. The lexical category of each word, Named Entities and Multi-Word Terms, were used to reformulate multiple questions into new individual questions, and then a Convolutional Neural Network was used to classify them, allowing to find or build adequate answers to improve the quality of the results, which is fundamental in QAsystems. A dataset with multiple questions was also created to evaluate our reformulation method, since it was not possible to find any. On the other hand, for the evaluation of the question classification model, we used the TREC, Simple Questions, Web Questions, Wiki Movies and Curated TREC datasets, translated into Spanish. Both tasks achieved promising results for further work.

Hybrid extraction of multi-word terms: an application on vibration-based condition monitoring technique

In this paper, we present an advanced domain-specific multi-word terminology extraction method. Our hybrid approach for automatic term identification benefits from both statistical and linguistic approaches. Our main goal is to reduce as much as possible the human effort in term selection tasks as well as to provide a wide-range and representative terminology of a domain. We emphasize in identification of verb or noun phrases multi-word terms, in neologisms and technical jargons. Our architecture applies the term frequency-inverse document frequency (TF-IDF) algorithm to a domain-specific textual corpus in order to measure a unit’s importance in it. We also use techniques to filter out nested terms of a candidate term taking into consideration its frequency by itself in the corpus. In addition, the exported terms are filtered out based on a stop-word list and linguistic criteria. To further reduce the size of the candidate terms and achieve accurate and precise terminologies, our method automatically validates them against a general-purpose corpus. Our study based on a small corpus of vibration-based condition monitoring domain shows that most extracted terms have nice correspondence to the domain of condition monitoring concepts and notions.

Term and translation variation of multiword terms

Phraseology is central to specialized language. In scientific and technical communication, multiword terms (MWTs) (e.g. volatile organic compound) are the most frequent type of phraseological units. Rendering them into another language is not an easy task due to their cognitive complexity, the proliferation of different forms, and their unsystematic representation in terminographic resources. This often results in a broad spectrum of translations for MWTs, leading to higher term variation as a result of their composition by two or more constituents. In this study we carried out a quantitative and qualitative analysis of English term variants of MWTs from the environmental domain and their translations into Spanish. The focus was on translation variation and its occurrence in different linguistic resources.

О ВАРИАНТНОСТИ ВЫРАЖЕНИЯ ПОДЧИНИТЕЛЬНЫХ ОТНОШЕНИЙ В АНАТОМИЧЕСКИХ ТЕРМИНАХ

The article reveals the patterns of using inconsistent and consistent attribute in the structure of the Latin anatomical multi-word term and in its equivalents in other languages. In search of criteria for choosing one or another method of determination, processes of lexicalization of a complex term, semantic analysis of components of a multi-word term are considered, contrast analysis is carried out to compare the Latin term with its equivalents in other languages. On the basis of the analysis of the International Anatomical Terminology, cases of variability of the same term with consistent and inconsistent attribute are considered. Professional language usage is considered as one of the reasons for the rare interchangeability of consistent and inconsistent attribute denoting the same anatomical object. Based on the theory of psychosystematics, the author concludes that there is a difference in the semantic perception of variants of the two-word term with consistent and inconsistent attribute in connection with the different thought mechanism of the incident in the formation of the noun and adjective.

Information retrieval from multi-domain specific research proposal using hierarchical-based neural network clustering algorithm

Research project selection is an essential task for government and private agencies. When a huge number of research proposals are received, it is common to group them along with their similarities in research discipline areas. In this paper, a new framework is proposed to assign the proposal to reviewer and identify the proposal theme automatically. First, a list of multi word terms are extracted automatically from the domain specific corpus of text documents into a hierarchy based on their semantic similarity. Second, the proximity of two vectors can be calculated using measures such as probabilistic correlation. The conditional probability is used to model the probability that a term appear together in a record. Third, a hierarchical-based neural network clustering algorithm is used to get an information retrieval from multi-domain specific research proposal.

INFORMATION RETRIEVAL FROM MULTI DOMAIN SPECIFIC RESEARCH PROPOSAL USING HIERARCHICAL BASED NEURAL NETWORK CLUSTERING ALGORITHM

Research project selection is an essential task for government and private agencies. When a huge number of research proposals are received, it is common to group them along with their similarities in research discipline areas. In this paper, a new framework is proposed to assign the proposal to reviewer and identify the proposal theme automatically. First, a list of multi word terms are extracted automatically from the domain specific corpus of text documents into a hierarchy based on their semantic similarity. Second, the proximity of two vectors can be calculated using measures such as probabilistic correlation. The conditional probability is used to model the probability that a term appear together in a record. Third, a hierarchical-based neural network clustering algorithm is used to get an information retrieval from multi-domain specific research proposal.

AUTOMATION OF SEARCHING FOR TERMS IN THE EXPLANATORY DICTIONARY

In this paper, an approach for automating the search for interpretations of terms for a specific domain in explanatory dictionary and on the Internet is proposed. A mathematical model of the explanatory dictionary is developed. It bases on the structure of the dictionary entry. The methodology for setting up an analyzer of a dictionary entry in a dictionary that has not been used before is developed. A methodology for automated search for one-word and multiword terms in electronic dictionaries has been developed. It bases on scanning dictionary entries in search for term using Regular Expressions. The automation of searching on the Internet resource using a browser automation tool Selenium is proposed. Automated analysis of search results in according to subject area have been developed in two methods. If there is a stylistic label in the structure of a dictionary entry, which indicates the area of the polysemantic term using, the results are corrected by filtering out definitions that do not correspond to this subject area. If there is no stylistic label, the search results are filtering out in the way of screening out definitions for occurrence of search terms. The creation of Dictionary bank for storing set-up to search electronic dictionaries is proposed. The program product, which allows search organizing in added and built-in electronic dictionaries and on the Internet resource, was developed. Using the program requires involvement of an expert to correct and verify the results. Working with the program requires the involvement of an expert to correct and verify the results. The effectiveness of the approach is confirmed experimentally. Groups of English, Russian and Ukrainian terms from different subject areas were used in the experiments. Formulas to determine the time spent on searching are proposed to assess the effectiveness of the implementation of the developed methods. The results showed a reduce of the time spent on search in the automated mode in about 5 times compared to the manual one. It is shown that adding an explanatory dictionary of a specialized subject area gives the most certain definition of terms in search process.