Examinando por Autor "Poibeau, Thierry"
Mostrando 1 - 3 de 3
Resultados por página
Opciones de ordenación
Publicación Automatic enjambment detection as a new source of evidence in Spanish versification(2017) Ruiz Fabo, Pablo; González-Blanco García, Elena; Poibeau, Thierry; Martínez Cantón, Clara IsabelWe present software to automatically identify enjambment (and its type) in Spanish. Traditionally, enjambment is described as a device whereby no syntactic pause occurs at the end of a poetry line, splitting a phrase across two lines. Most definitions highlight the conflict between syntactic and metrical units, yielding stylistic effects like double interpretations. In Spanish versification, Quilis (1964) performed poetry reading experiments, characterizing enjambment as happening only when very cohesive syntactic units, which it would be unnatural to interrupt, are split across lines. Spang (1983) noted that splitting verbs and their subject or object across two lines also triggers (softer) enjambment-like effects. These characterizations are still considered current, but some points in them are debated. To systematically gather evidence on enjambment, we created a Natural Language Processing-based system that automatically detects and types enjambments as characterized above. For evaluation, we manually annotated a reference corpus. We consider system results satisfactory; F1 varied depending on enjambment type and poems’ period. A system and corpus description, and evaluation are at: https://sites.google.com/site/spanishenjambment/ We are not aware of large-sample enjambment studies across periods, literary movements, or versification types in Spanish, or other languages. Automatic detection can provide quantitative evidence for questions in verse theory, e.g.: To what an extent is enjambment used differently in free verse vs. traditional versification? Applying the system to 3750 sonnets covering four centuries is shedding light on unclear points in the definition of enjambment. The system finds line-pairs formally fitting the description of enjambment, but that, upon human validation, we’d consider borderline cases, given other stylistic factors, e.g. hyperbaton. Conversely, our annotators are sometimes surprised that certain line-pairs are not considered enjambed in the typology. Automatic identification of many examples, plus expert validation, is helping towards a more nuanced redefinition of enjambment.Publicación Distant Rhythm: Automatic Enjambment Detection on Four Centuries of Spanish Sonnets(2017) Ruiz Fabo, Pablo; Poibeau, Thierry; Martínez Cantón, Clara IsabelEnjambment takes place when a syntactic unit is broken up across two lines of poetry, giving rise to different stylistic effects. In Spanish literary studies, detailed case-studies of the phenomenon based on single authors exist. However, a larger-scale study spanning hundreds of major and minor authors, across several centuries, is not available so far. Towards that need, we have developed software based on Natural Language Processing (NLP), to automatically identify enjambment (and its type) in Spanish. To evaluate the system, we manually annotated two reference corpora (one diachronic, one from the 20th century). Results are satisfactory for the system's first version, with F1 varying depending on period and enjambment type. As a scholarly corpus to apply the tool, from public HTML sources we created a diachronic corpus covering four centuries of sonnets (3750 poems). We applied the tool to analyze the occurrence of enjambment across stanzaic boundaries in different periods.Publicación Enjambment Detection in a Large Diachronic Corpus of Spanish Sonnets(Association for Computational Linguistics, 2017) Ruiz Fabo, Pablo; Poibeau, Thierry; González-Blanco García, Elena; Martínez Cantón, Clara IsabelEnjambment takes place when a syntactic unit is broken up across two lines of poetry, giving rise to different stylistic effects. In Spanish literary studies, there are unclear points about the types of stylistic effects that can arise, and under which linguistic conditions. To systematically gather evidence about this, we developed a system to automatically identify enjambment (and its type) in Spanish. For evaluation, we manually annotated a reference corpus covering different periods. As a scholarly corpus to apply the tool, from public HTML sources we created a diachronic corpus covering four centuries of sonnets (3750 poems), and we analyzed the occurrence of enjambment across stanzaic boundaries in different periods. Besides, we found examples that highlight limitations in current definitions of enjambment.