94 research outputs found
English as a lingua franca in academic publishing: using round-trip translation to estimate linguistic revision difficulty
No full textIn questo articolo viene presentato un metodo innovativo per stimare la difficoltà di revisione di testi accademici da pubblicare in inglese come lingua franca, allo scopo di ottimizzare il lavoro e le performance dei revisori. In questo modo la complessa questione degli standard linguistici nella produzione accademica in inglese viene affrontata da una nuova prospettiva. Il metodo round-trip translation (RTT) che consiste nel tradurre automaticamente un testo in un’altra lingua e poi di nuovo nella lingua di partenza, viene applicato in combinazione con BLEU, un metodo automatico per stimare la somiglianza dei testi tradizionalmente utilizzato nella valutazione della traduzione automatica. Calcoliamo la somiglianza tra il manoscritto e il prodotto della rispettiva RTT per ottenere una valutazione a priori della difficoltà di revisione e testiamo l’efficacia del nostro metodo. Insieme alla ricerca sperimentale vengono discusse alcune questioni teoriche riguardanti l’inglese come lingua franca e la pratica della revisione linguistica nel contesto dell’editoria accademica internazionale. I risultati degli esperimenti suggeriscono che il metodo RTT può essere utilizzato con successo per stimare la difficoltà di revisione. La varietà dei dati utilizzati negli esperimenti dimostra l’applicabilità del metodo al di là dei casi oggetto di studio. Concludiamo con una discussione delle limitazioni del lavoro e di possibili sviluppi futuri.In this paper the feasibility of a novel method for estimating revision difficulty of academic texts to be published in English as a lingua franca (ELF) is examined with the purpose of optimizing revisors’ workflow and performance. In doing so, we also address the complex issue of linguistic standards in ELF academic productions from a new perspective. The method of automatic round-trip translation (RTT), which consists in machine-translating a text into another language and back into the source language, is applied in conjunction with BLEU, an automatic text similarity method traditionally used in machine translation evaluation. The similarity between a manuscript and its round-trip translation is calculated to obtain a priori evaluation of revision difficulty, and the efficacy and reliability of the method are tested. Alongside with the experimental research, some theoretical issues regarding ELF and the practice of revision in the hyperglobalized context of international academic publishing are discussed. The results of the experiments show that the RTT-based method can be successfully used to estimate revision difficulty. The variety of data used in the experiments demonstrates the applicability of the method beyond the cases under study. Some limitations and further developments of the study are also briefly discussed
Forliviamo
No full textForliviamo è una app sviluppata nell'ambito del progetto CALL-ER - Context-Aware Language Learning in Emilia-Romagna: turismo esperienziale e apprendimento nelle città del Multicampus UNIBO realizzato dall'Università di Bologna, Campus di Forlì, (Alte competenze per la ricerca e il trasferimento tecnologico POR FSE 2014/2020, Risorse Umane per la Specializzazione Intelligente, Regione Emilia-Romagna), grazie anche all’esperienza maturata nell’ambito del progetto europeo Ilocalapp per lo sviluppo della app UniON! Bologna.
La app promuove la conoscenza dei luoghi della provincia di Forlì-Cesena (città del multicampus universitario UNIBO) e, contestualmente, favorisce l’apprendimento dell’italiano attraverso percorsi linguistico-culturali di tipo esperienziale. Particolare interesse è rivolto agli spazi urbani e naturalistici riqualificati, fuori dalle rotte tradizionali, grazie anche ad una stretta collaborazione con le associazioni del territorio
!Translate : When You Cannot Cook Up a Translation, Explain
No full textIn the domain of cuisine, both dishes and in- gredients tend to be heavily rooted in the local context they belong to. As a result, the asso- ciated terms are often realia tied to specific cultures and languages. This causes difficulties for non-speakers of the local language and ma- chine translation (MT) systems alike, as it im- plies a lack of the concept and/or of a plausible translation. MT typically opts for one of two alternatives: keeping the source language terms untranslated or relying on a hyperonym/near- synonym in the target language, provided one exists. !Translate proposes a better alterna- tive: explaining. Given a cuisine entry such as a restaurant menu item, we identify culture- specific terms and enrich the output of the MT system with automatically retrieved definitions of the non-translatable terms in the target lan- guage, making the translation more actionable for the final user
AriEmozione 1.0
No full textThe corpus AriEmozione 1.0 contains a selection of operas composed between 1655 and 1765, with each verse annotated (labeled) with an emotion.
AriEmozione 1.0 is a subset of the materials collected by project CORAGO.
We consider only the lyrical text of the arias in each of the 678 operas included. All texts are written in Italian of that period and articulated in verses.
The following classes have been used to identify emotions:
Amore (Love)
Gioia (Joy)
Ammirazione (Admiration)
Rabbia (Anger)
Tristezza (Sadness)
Paura (Fear)
Nessuna (Nothing)
The corpus is composed of training (1973 verses), development (250 verses) and testing (250 verses) partitions. Data partitioning made use of Kullback–Leibler divergence to create partitions with similar probability distribution.
Each partition tsv file contains the following columns:
Verse ID : unique aria and verse ID
Verse text: the text of the verse in the aria
Emotion: one of the six emotions described above, or Nessuna.
Confidence: expression of confidence by the annotators, with the possible values of "Totalmente sicura" (very confident), "Abbastanza sicura" (somewhat confident), "Forti dubbi" (very doubtful).
How to cite:
@inproceedings{fernicola2020ariemozione,
title={AriEmozione: Identifying Emotions in Opera Verses},
author={Fernicola, Francesco and Zhang, Shibingfeng and Garcea, Federico and Bonora, Paolo and Barr{\'o}n-Cede\~no, Alberto},
booktitle={Italian Conference on Computational Linguistics},
year={2020}
AriEmozione 2.0: Identifying Emotions in Opera Verses and Arias
No full textWe present the task of identifying the emotions conveyed by the lyrics of Italian opera arias. We shape the task as a multi-class supervised problem, considering the six emotions from Parrot’s tree: love, joy, admiration, anger, sadness, and fear. We manually annotated an opera corpus with 2.5k instances at the verse level and experimented with different classification models and representations to identify the expressed emotions. Our best-performing models consider character 3-gram representations and reach relatively low levels of macro-averaged F1. Such performance reflects the difficulty of the task at hand, partially caused by the size and nature of the corpus: relatively short verses written in 18th-century Italian. Building on what we learned from the verse-level setting, we adopt a higher granularity and increase the size of the corpus. First, we switch from verses to arias in order to have longer and more expressive texts. Second, we construct a new corpus with 40k arias (∼
90k verses). This new dataset contains silver data, annotated by self-learning on the basis of an ensemble of binary classifiers.
We then experiment with more sophisticated representations, by learning an embedding space and using it to train new models for the identification of emotions at the aria level, obtaining a significant performance boost
AriEmozione: identifying emotions in opera verses
No full textWe present a new task: the identification of the emotions transmitted in Italian opera arias at the verse level. This is a relevant problem for the organization of the vast repertoire of Italian Opera arias available and to enable further analyses by both musicologists and the lay public. We shape the task as a multi-class supervised problem, considering six emotions: love, joy, admiration, anger, sadness, and fear. In order to address it, we manually-annotated an opera corpus with 2.5k verses —which we release to the research community— and experimented with different classification models and representations. Our best-performing models reach macro-averaged F1 measures of ~0.45, always considering character 3-grams representations. Such performance reflects the difficulty of the task at hand, partially caused by the size and nature of the corpus, which consists of relatively short verses written in 18th-century Italian
Centro in Periferia. Papiri, ostraka e tasselli di lingua latina per una literacy d'Oriente
No full textLa declinatio nel commento serviano a Virgilio
No full textThe paper takes into account the evidences dealing with the declinatio in Servian commentary on Virgil's works in order to provide an overview of Servius' grammatical theory in this respect and of the analogies and differences with the authors of Artes
Recrutement et intégration des nouveaux arrivants de langue française à l’Île-du-Prince-Édouard
No full textLa distribution régionale des immigrants est d’une importance capitale au Canada. L’enjeu est d’assurer un certain équilibre au niveau de la croissance démographique et économique pour chaque province et territoire (Garcea et Golebioska, 2012 : 72). À l’intérieur des initiatives d’établissement de nouveaux arrivants en région au Canada adoptées pendant les dix dernières années, on retrouve une politique qui vise «à accroître le nombre d’immigrants francophones souhaitant s’établir dans une des communautés francophones hors Québec» (Garcea et Golebioska, 2012 : 78). Dans ce sens, le désir du gouvernement fédéral et des gouvernements provinciaux de voir de nouveaux arrivants maîtrisant l’une des deux langues officielles du Canada s’installer dans une communauté de langue officielle en situation minoritaire (CLOSM) afin de contribuer à renverser, ou à tout le moins ralentir, le processus de décroissance démographique de ces communautés est palpable
SIC SERVIVS MAGISTER EXPOSVIT: l’auctoritas mitografica di Servio e le interconnessioni fra i commentatori tardi
No full text- …
