Neural machine translation systems have revolutionized translation processes in terms of quantity and speed in recent years, and they have even been claimed to achieve human parity. However, the ...quality of their output has also raised serious doubts and concerns, such as loss in lexical variation, evidence of “machine translationese”, and its effect on post-editing, which results in “post-editese”. In this study, we analyze the outputs of three English to Slovenian machine translation systems in terms of lexical diversity in three different genres. Using both quantitative and qualitative methods, we analyze one statistical and two neural systems, and we compare them to a human reference translation. Our quantitative analyses based on lexical diversity metrics show diverging results; however, translation systems, particularly neural ones, mostly exhibit larger lexical diversity than their human counterparts. Nevertheless, a qualitative method shows that these quantitative results are not always a reliable tool to assess true lexical diversity and that a lot of lexical “creativity”, especially by neural translation systems, is often unreliable, inconsistent, and misguided.
Poročilo o 34. evropski poletni šoli logike, jezika in informatike (European Summer School of Logic, Language and Information (ESSLLI)), ki je potekala med 31. julijem in 11. avgustom 2023 na ...Fakulteti za računalništvo in informatiko v Ljubljani.
Abstract The COVID pandemic spurred the use of various metaphors, some very common and universal, others depending on the language, country and culture. The use of metaphors by the general public, ...especially in languages other than English, has not yet been sufficiently investigated, one of the reasons being the lack of resources and automatic tools for metaphor analysis. To fill this gap, we introduce TCMeta, a dataset of tweets annotated for metaphors around COVID-19, in two languages from ten different countries. The dataset contains metaphoric phrases covering four source domains. Furthermore, we introduce a semi-automatic methodology to annotate more than 2000 tweets in English and Slovene. To the best of our knowledge, this is the first multilingual semi-automatically compiled dataset of user-generated texts aimed at investigating metaphorical language about the pandemic. It is also the first Slovene dataset of tweets annotated for metaphors.
Dehumanisation involves the perception and or treatment of a social group's members as less than human. This phenomenon is rarely addressed with computational linguistic techniques. We adapt a ...recently proposed approach for English, making it easier to transfer to other languages and to evaluate, introducing a new sentiment resource, the use of zero-shot cross-lingual valence and arousal detection, and a new method for statistical significance testing. We then apply it to study attitudes to migration expressed in Slovene newspapers, to examine changes in the Slovene discourse on migration between the 2015-16 migration crisis following the war in Syria and the 2022-23 period following the war in Ukraine. We find that while this discourse became more negative and more intense over time, it is less dehumanising when specifically addressing Ukrainian migrants compared to others.
Being secondary plant metabolites, polyphenols represent a large and diverse group of substances abundantly present in a majority of fruits, herbs and vegetables. The current contribution is focused ...on their bioavailability, antioxidative and anticarcinogenic properties. An overview of extraction methods is also given, with supercritical fluid extraction highlighted as a promising eco-friendly alternative providing exceptional separation and protection from degradation of unstable polyphenols. The protective role of polyphenols against reactive oxygen and nitrogen species, UV light, plant pathogens, parasites and predators results in several beneficial biological activities giving rise to prophylaxis or possibly even to a cure for several prevailing human diseases, especially various cancer types. Omnipresence, specificity of the response and the absence of or low toxicity are crucial advantages of polyphenols as anticancer agents. The main problem represents their low bioavailability and rapid metabolism. One of the promising solutions lies in nanoformulation of polyphenols that prevents their degradation and thus enables significantly higher concentrations to reach the target cells. Another, more practiced, solution is the use of mixtures of various polyphenols that bring synergistic effects, resulting in lowering of the required therapeutic dose and in multitargeted action. The combination of polyphenols with existing drugs and therapies also shows promising results and significantly reduces their toxicity.