1,721,226 research outputs found
Proceedings of the Workshop on Multi-word Units in Machine Translation and Translation Technology (MUMTTT 2015)
No full textThis volume documents the proceedings of the 2nd Workshop on Multi-word Units in Machine Translation and Translation Technology (MUMTTT 2015), held on 1-2 July 2015 as part of the EUROPHRAS 2015 conference: "Computerised and Corpus-based Approaches to Phraseology: Monolingual and Multilingual Perspectives" (Málaga, 29 June – 1 July 2015). The workshop was sponsored by European COST Action PARSing and Multi-word Expressions (PARSEME) under the auspices of the European Society of Phraseology (EUROPHRAS), the Special Interest Group on the Lexicon of the Association for Computational Linguistics (SIGLEX), and SIGLEX's Multiword Expressions Section (SIGLEX-MWE). The workshop was co-chaired by Gloria Corpas Pastor (Universidad de Málaga), Ruslan Mitkov (University of Wolverhampton), Johanna Monti (Università degli Studi di Sassari), and Violeta Seretan (Université de Genève). It received the support of the Advisory Board, composed of Dmitrij O. Dobrovol'skij (Russian Academy of Sciences, Moscow), Kathrin Steyer (Institut für Deutsche Sprache, Mannheim), Agata Savary (Université François Rabelais Tours), Michael Rosner (University of Malta), and Carlos Ramisch (Aix-Marseille Université).
The topic of the workshop was the integration of multi-word units in machine translation and translation technology tools. In spite of the recent progress achieved in machine translation and translation technology, the identification, interpretation and translation of multi-word units still represent open challenges, both from a theoretical and from a practical point of view. The idiosyncratic morpho-syntactic, semantic and translational properties of multi-word units poses many obstacles even to human translators, mainly because of intrinsic ambiguities, structural and lexical asymmetries between languages, and, finally, cultural differences. After a successful first edition held in Nice on 3 September 2013 as part of the Machine Translation Summit XIV, the present edition provided a forum for researchers working in the fields of Linguistics, Computational Linguistics, Translation Studies and Computational Phraseology to discuss recent advances in the area of multi-word unit processing and to coordinate research efforts across disciplines.
The workshop was attended by 53 representatives of academic and industrial organisations. The programme included 11 oral and 4 poster presentations, and featured an invited talk by Kathrin Steyer, President of EUROPHRAS. We received 23 submissions, hence the MUMTTT 2015 acceptance rate was 65.2%. The papers accepted are indicative of the current efforts of researchers and developers who are actively engaged in improving the state of the art of multi-word unit translation
Evaluating Large Language Models in Relationship Extraction from Unstructured Data: Empirical Study from Holocaust Testimonies
No full textRelationship extraction from unstructured data remains one of the most challenging tasks in the field of Natural Language Processing (NLP). The complexity of relationship extraction arises from the need to comprehend the underlying semantics, syntactic structures, and contextual dependencies within the text. Unstructured data poses challenges with diverse linguistic patterns, implicit relationships, contextual nuances, complicating accurate relationship identification and extraction. The emergence of Large Language Models (LLMs), such as GPT (Generative Pre-trained Transformer), has indeed marked a significant advancement in the field of NLP. In this work, we assess and evaluate the effectiveness of LLMs in relationship extraction in the Holocaust testimonies within the context of the Historical realm. By delving into this domain specific context, we aim to gain deeper insights into the performance and capabilities of LLMs in accurately capturing and extracting relationships within the Holocaust domain by developing a novel knowledge graph to visualise the relationships of the Holocaust. To the best of our knowledge, there is no existing study which discusses relationship extraction in Holocaust testimonies. The majority of current approaches for Information Extraction (IE) in historic documents are either manual or Optical Character Recognition (OCR) based. Moreover, in this study, we found that the Subject-Object-Verb extraction using GPT3- based relations produced more meaningful results compared to the Semantic Role labeling based triple extraction
Preliminary evaluation of ChatGPT as a machine translation engine and as an automatic post-editor of raw machine translation output from other machine translation engines
Full text linkThis preliminary study consisted of two experiments. The first aimed to gauge the translation quality obtained from the free-plan version of ChatGPT in comparison with the free versions of DeepL Translator and Google Translate through human evaluation, and the second consisted of using the free-plan version of ChatGPT as an automatic post-editor of raw output from the pay-for version of DeepL Translator (both monolingual and bilingual full machine translation post-editing). The experiments were limited to a single language pair (from English to Italian) and only one text genre (Wikipedia articles).
In the first experiment, DeepL Translator was judged to have performed best, Google Translate came second, and ChatGPT, last.
In the second experiment, the free-plan version of ChatGPT equalled average human translation (HT) levels of lexical variety in automatic monolingual machine translation post-editing (MTPE) and exceeded average HT lexical variety
levels in automatic bilingual MTPE. However, only one MT marker was considered, and the results of the post-editing were not quality-assessed for other features of MTPE that distinguish it from HT. It would therefore be unadvisable to generalize these findings at present.
The author intends to carry out new translation experiments during the next academic year with ChatGPT Plus, instead of the free-plan version, both as an MT engine and as an automatic post-editor. The plan is to continue to evaluate the results manually and not automatically
Author gender identification for Urdu articles
No full textThis is an accepted manuscript of an article published by Springer in Lecture Notes in Computer Science on 21/09/2022. The accepted version of the publication may differ from the final published versionIn recent years, author gender identification has gained considerable attention in the fields of computational linguistics and artificial intelligence. This task has been extensively investigated for resource-rich languages such as English and Spanish. However, researchers have not paid enough attention to perform this task for Urdu articles. Firstly, I created a new Urdu corpus to perform the author gender identification task. I then extracted two types of features from each article including the most frequent 600 multi-word expressions and the most frequent 300 words. After I completed the corpus creation and features
extraction processes, I performed the features concatenation process. As a result each article was represented in a 900D feature space. Finally, I applied 10 different well-known classifiers to these features to perform the author gender identification task and compared their performances against state-of-the-art pre-trained multilingual language models, such as mBERT, DistilBERT, XLM-RoBERTa and multilingual DeBERTa, as well as Convolutional Neural Networks (CNN). I conducted extensive experimental studies which show that (i) using the most frequent 600 multi-word expressions as features and concatenating them with the most frequent 300 words as features improves the accuracy of the author gender identification task, and (ii) support vector machines outperforms other classifiers, as well as fine-tuned pre-trained language models and CNN. The code base and the corpus can be found at: https://github.com/raheem23/Gender_Identification_Urdu
Current evidence of post-editese: differences between post-edited neural machine translation output and human translation revealed through human evaluation
Full text linkThe experiment reported in this paper is a follow-up to one conducted in 2017/2018. The new experiment aimed to establish if the previously observed lexical impoverishment in machine translation post-editing (MTPE) has become more marked as technology has developed or if it has attenuated. This was done
by focusing on two n-grams, which had been previously identified as MT markers, i.e., n-grams that give rise to translation solutions that occur with a higher frequency in MTPE than is natural in HT. The new findings suggest that lexical impoverishment in the two short texts examined has indeed diminished with DeepL Translator.
The new experiment also considered possible syntactic differences, namely the number of text segments in the target text. However no significant difference waThe experiment reported in this paper is a follow-up to one conducted in 2017/2018. The new experiment aimed to establish if the previously observed lexical impoverishment in machine translation post-editing (MTPE) has become more marked as technology has developed or if it has attenuated. This was done by focusing on two n-grams, which had been previously identified as MT markers, i.e., n-grams that give rise to translation solutions that occur with a higher frequency in MTPE than is natural in HT. The new findings suggest that lexical impoverishment in the two short texts examined has indeed diminished with DeepL Translator.
The new experiment also considered possible syntactic differences, namely the number of text segments in the target text. However no significant difference was observed.
The participants were asked to complete a short questionnaire on how they went about their tasks. It emerged that it was helpful to consult the source language text while post-editing, and the original unedited raw output while self-revising, suggesting that monolingual MTPE of the two chosen texts would have been unwise.
Despite not being given specific guidelines, the productivity of the post-editors increased. If the ISO 18587:2017 recommendation of using as much of the MT output as possible had been strictly followed, the MTPE would have been easier to distinguish from HT. If this can be taken to be generally true, it suggests that it is neither necessary nor advisable to follow this recommendation when lexical diversity is crucial for making the translation more engaging
Do translators use machine translation and if so, how? Results of a survey held among professional translators
Full text linkThe author conducted an anonymous online survey between 23 July and 21 October 2022 to gain insight into the proportion of translators that use machine translation (MT) in their translation workflow and the various ways they do. The results show that translators with more experience are less likely to accept MT post-editing (MTPE) assignments than their less experienced colleagues but are equally likely to use MT themselves in their translation work. Translators who deal with lower-resource languages are also less likely to accept MTPE jobs, but there is no such relationship regarding the use of MT in their own workflow. When left to their own devices, only 18.57% of the 69.54% of respondents that declared that they use MT while translating always or usually use it in the way the pioneers of MT envisaged, i.e., MTPE. Most either usually or always prefer to use MT in a whole range of other ways, including enabling MT functions in CAT tools and doing hybrid post-editing; using MT engines as if they were dictionaries; and using MT for inspiration. The vast majority of MT-users see MT as just another tool that their clients do not necessarily need to be informed about
When Multiwords Go Bad in Machine Translation
No full textThis paper addresses the impact of multiword translation errors in machine translation (MT). We have analysed translations of multiwords in the OpenLogos rule-based system (RBMT) and in the Google Translate statistical system (SMT) for the English-French, English-Italian, and English-Portuguese language pairs.
Our study shows that, for distinct reasons, multiwords remain a problematic area for MT independently of the approach, and require adequate linguistic quality evaluation metrics founded on a systematic categorization of errors by MT expert linguists.
We propose an empirically-driven taxonomy for multiwords, and highlight the need for the development of specific
corpora for multiword evaluation. Finally, the paper presents the Logos approach to multiword processing, illustrating how semantico-syntactic rules contribute to multiword translation quality
Multiword units translation evaluation in machine translation: another pain in the neck?
No full textRecent studies have highlighted that the translation of Multiword Units (MWUs) by Machine Translation (MT) is still an open challenge, whatever the adopted approach (statistical, rule-based or example- based). The difficulties in translating automatically this recurrent, complex and varied lexical phenomenon originate from its lexical, syntactic, semantic, pragmatic and/or statistical but also translational idiomaticity. It is widely acknowledged that in order to achieve significant improvements in Machine Translation and translation technologies it is important to develop resources, which can be used both for Statistical Machine Translation (SMT) training and evaluation purposes. There is therefore, the need to develop linguistic re- sources, mainly parallel corpora annotated with MWUs which can help improve the MT quality in particular as regards translation of MWUs in context and discontinuous MWUs. In this paper, we analyse the state of the art concerning MWU-aware MT evaluation metrics, the availability of both benchmarking resources and annotation guidelines and procedures
Ontologies in the context of knowledge organization and interoperability in e-government services
Full text linkA philosophical and technical approach of the concept of ontology is introduced and the importance of ontologies to structure, manage and retrieve the knowledge of different scientific fields and various political and social contexts is enhanced. The tools related to ontology management and the languages used for ontology creation are revised. In particular, an identification of the web-based ontology mark-up languages that contribute to knowledge representation and organization in this electronic and hypertextual environment, such as DAML+OIL, OWL and OML, is carried out. Finally, ontologies are presented as the basis for semantic web development and as a tool to guarantee information interoperability in e-government services. In this sense, some examples of initiatives for ontology application in electronic public services are provided
Language Resources for Italian: towards the Development of a Corpus of Annotated Italian Multiword Expressions
Full text linkQuesto contributo descrive la prima risorsa italiana annotatata con polirematiche. Sono state preparate due versioni del dataset: la prima con una
lista di polirematiche senza contesto, e la seconda con annotazione in contesto.
Il contributo discute le problematiche emerse durante l’annotazione e riporta il grado di accordo tra annotatori per entrambi i tipi di annotazione. Infine vengono presentati i risultati del primo impiego della nuova risorsa, ovvero l’estrazione automatica di polirematiche per l’italiano.This paper describes the first resource annotated for multiword expressions (MWEs) in Italian. Two versions of this dataset have been prepared: the first with a fast markup list of out-of-context MWEs, and the second with an in-context annotation, where the MWEs are entered with their contexts. The paper also discusses annotation issues and reports the inter-annotator agreement for both types of annotations. Finally, the results of the first exploitation of the new resource, namely the automatic extraction of Italian MWEs, are presented
- …
