Kalbotyra
Not a member yet
2154 research outputs found
Sort by
How to annotate a corpus for research on frames in the language of mathematics
Get PDFThis paper outlines a plan for designing a corpus-linguistic project from which an analysis of frames specific to mathematical proof texts shall be derived. Previous work has already developed instances of frames for mathematical texts. We have argued that these frames are well suited to model how background knowledge enriches explicitly given information. To this end, a collection of mathematical texts needs to be annotated.We describe the idea behind the corpus annotation, frame semantics and how we adapted frames for mathematical texts in particular by distinguishing between structural proof frames and ontological ones. Ontological frames for instance correspond to proof techniques, while structural frames model domain knowledge like exact definitions of mathematical structures. We describe annotation principles, explain qualifications annotators should have and how such annotations can be evaluated.We explain potential linguistic research questions: The corpus will allow us to study the linguistic means by which frames are introduced and signaled. We plan to generalize beyond previous case-based analyses and to provide a foundation for broader empirical research on mathematical language and practice. We argue that these can be used in deeper semantic parsing and the development of interactive theorem-proving software. We furthermore assume that this perspective on mathematical text will also have implications on the philosophy of mathematics and its practice
Daiktavardžio pyktis valentingumas Dabartinės lietuvių kalbos tekstyne
Get PDFBy using methods of interpretation, descriptive analysis, grammatical analysis, semantic and lexical analysis, 292 sentences containing different forms of the noun pyktis ʻangerʼ drawn from the Corpus of Contemporary Lithuanian Language were analysed. The research showed that the syntactic valency of the noun pyktis ʻangerʼ partly differs from that of the verb pykti(s) ‘to be angry (with)ʼ. The findings indicate that the syntactic valency of the noun pyktis ʻangerʼ is influenced by the semantics and grammar of the verb pykti(s) ‘to be angryʼ and nouns within its minimal semantic field. Specifically, the nominative case governed by the verb pykti(s) ‘to be angryʼ is replaced by the genitive case governed by the noun pyktis ʻangerʼ. Similar to the verb pykti(s) ‘to be angryʼ, the noun pyktis ʻangerʼ requires constructions such as ant ʻonʼ + genitive, dėl ʻbecause ofʼ + genitive, už ʻforʼ + accusative, su ʻwithʼ + instrumental and a subordinate clause. Furthermore, pyktis ʻangerʼ, like priešiškas ʻhostileʼ, jausmas ʻfeelingʼ, nusistatymas ʻattitudeʼ and piktumas ʻmaliceʼ, requires a dative actant and can take a construction with atžvilgiu ʻin relation toʼ instead of the dative. Similar to nusistatymas ʻattitudeʼ, pyktis ʻangerʼ can take a construction with prieš ʻagainstʼ and tarp ʻamongʼ.The noun pyktis ʻangerʼ has the same semantic valency as the verb pykti(s) ʻto be angryʼ, just there are more ways to express the contentive and the percipient. When a person, who is experiencing anger, is mentioned or emphasized in a sentence, the noun pyktis ʻangerʼ agrees with the genitive or is used in a construction with tarp ʻamongʼ percipient. When the noun pyktis ʻangerʼ agrees with a dative or is involved in constructions with ant ʻonʼ, prieš ʻagainstʼ or atžvilgiu ʻin relation toʼ contentive, it expresses a content of state. When the cause is actualized in a sentence, pyktis ʻangerʼ agrees with the construction dėl ʻbecause ofʼ + genitive causative, whereas, when the reason is highlighted, it is used with the construction už ʻforʼ + accusative motive. However, it can describe a peripheral person, in this case pyktis ʻangerʼ agrees with the construction su ʻwithʼ + instrumental comitative.The collected examples show that the semantic valency of pyktis ʻangerʼ is more diverse than that of the verb. A genitive actant can indicate not only humans, but objects or abstract concepts too. An anger can be directed towards anything: humans, animate and inanimate objects, or abstract concepts. Nonetheless, a construction su ʻwithʼ + instrumental is not unique to humans, it can indicate abstract concepts too.Hence the noun pyktis ʻangerʼ inherits the valency of verbs pykti(s) ʻto be angryʼ, while also acquiring new properties from nouns within its minimal semantic field.Straipsnyje nagrinėjamas daiktavardžio pyktis valentingumas. Remiantis Dabartinės lietuvių kalbos tekstyno medžiaga norėta nustatyti šio daiktavardžio junglumą trimis aspektais: sintaksiniu, semantiniu bei leksiniu. Bandyta išsiaiškinti, kas daro įtaką daiktavardžio pyktis valentingumui, kokias veiksmažodžio pykti(s) valentingumo savybes perima šis daiktavardis. Darbe taikyti interpretacijos, aprašomasis analitinis, gramatinės, semantinės, leksinės analizės metodai. Nustatyta, kad daiktavardis pyktis ne tik paveldi veiksmažodžių pykti ir pyktis valentingumą, bet ir įgyja papildomų savybių
KI als Hilfsmittel für die Formulierung von Suchanfragen in Korpora
Get PDFThe importance of corpus work and corpus-linguistic methods is steadily increasing in linguistic research and in the field of German as a Foreign Language (GFL), as they enable a comprehensive and systematic analysis of linguistic phenomena. Researchers often work with a large number of corpora that differ in their datasets and structures. One challenge in this work, however, is the diversity of query systems used across individual corpora. These systems can be complex and heterogeneous, which often requires a considerable amount of time to understand their functionality and use them effectively. This diversity can make the search for specific linguistic patterns or phenomena difficult and frustrating for both experienced users and beginners, as a great deal of time must be spent familiarizing oneself with the particularities of individual query systems. In response to these challenges, the integration of AI as a tool for formulating search queries in corpora may play a decisive role in the future.In this context, the present empirical contribution examines the potential of AI-supported systems for formulating complex search queries in COSMAS II. The study draws on 50 typical corpus-linguistic search tasks of different types, formulated by inexperienced users without prior knowledge of computational linguistics. The AI-generated queries are analyzed with regard to their formal correctness, precision, and functionality.Methodologically, the study is based on a qualitative analysis of the query results. The results indicate the conditions under which AI can be used as a supportive tool in corpus querying and where its limitations lie. The contribution also discusses what information needs to be available to the AI in order to effectively support corpus-linguistic search processes.Die Bedeutung von Korpusarbeit und korpuslinguistischen Methoden nimmt in der linguistischen Forschung und im DaF-Unterricht stetig zu, da sie eine umfassende und systematische Analyse sprachlicher Phänomene ermöglichen. Häufig arbeiten Forscherinnen und Forscher dabei mit einer Vielzahl von Korpora, die unterschiedliche Datensätze und Strukturen aufweisen. Eine zentrale Herausforderung besteht in der Vielfalt der jeweils eingesetzten Abfragesysteme. Diese sind oft komplex und heterogen, sodass ein erheblicher Zeitaufwand erforderlich ist, um ihre Funktionsweise zu verstehen und sie effektiv zu nutzen. Die damit verbundene Vielfalt kann die Suche nach spezifischen sprachlichen Mustern oder Phänomenen sowohl für erfahrene Nutzerinnen und Nutzer als auch für Anfängerinnen und Anfänger erschweren und frustrierend machen, da umfangreiche Einarbeitungszeiten in die jeweiligen Abfragesprachen notwendig sind. Angesichts dieser Herausforderungen kann die Integration von KI als Werkzeug zur Formulierung von Suchanfragen in Korpora künftig eine entscheidende Rolle spielen.Vor diesem Hintergrund untersucht der vorliegende empirische Beitrag das Potenzial KI-gestützter Systeme zur Formulierung Suchanfragen in COSMAS II. Grundlage der Untersuchung bilden 50 typische korpuslinguistische Suchaufgaben unterschiedlicher Art, die von unerfahrenen Nutzerinnen und Nutzern ohne computerlinguistische Vorkenntnisse formuliert werden. Die KI-generierten Abfragen werden hinsichtlich ihrer formalen Korrektheit, Präzision und Funktionalität analysiert.Methodisch basiert die Studie auf einer qualitativen Auswertung der Abfrageergebnisse. Die Ergebnisse zeigen, unter welchen Bedingungen KI als unterstützendes Werkzeug bei der Korpusabfrage eingesetzt werden kann und wo ihre Grenzen liegen. Darüber hinaus wird diskutiert, welche Informationen der KI zur Verfügung stehen müssen, um korpuslinguistische Suchprozesse effektiv zu unterstützen
Melanie Andresen. Computerlinguistische Methoden für die Digital Humanities: Eine Einführung für Geisteswissenschaftler:innen
Get PDFVorbemerkungen
„Die Digital Humanities ersetzen – selbstverständlich – die Geisteswissenschaften nicht, sondern erweitern ihr Methodenarsenal, und zwar dann, wenn Forschungsfragen durch empirische Arbeit mit großen Datensammlungen bearbeitet werden können“ (Krämer 2025, 11). Diese Aussage steht zu Beginn eines Essays, das den markanten Titel „Der Stachel des Digitalen: Geisteswissenschaften und Digital Humanities“ trägt. Darin setzt sich die Philosophieprofessorin Sybille Krämer kritisch mit dem Selbstbild der Geisteswissenschaften auseinander und zeigt auf, in welchem Verhältnis diese zu den Methoden der Digital Humanities stehen. Doch um welche Methoden handelt es sich dabei? Und welche Vorteile können sich daraus ergeben, wenn traditionelle Verfahren der Textanalyse, d. h. Textinterpretation und Hermeneutik, eine Erweiterung ihres „Methodenarsenals“ erfahren? In dem Studienbuch von Melanie Andresen, das hier zur Rezension steht, werden diese Fragen nicht nur auf theoretischer Ebene erörtert; es wird eine praxisorientierte Einführung präsentiert, die im Unterschied zu ähnlich gelagerten Überblicksdarstellungen (z. B. Alpaydin 2022; Mitkov 2022) den Schwerpunkt explizit auf die Anwendung computerlinguistischer Methoden in den Geisteswissenschaften legt
Building and annotating a bachelor paper abstract corpus: First findings about Latvian student abstracts
Get PDFThe paper discusses the problems identified in the early stages of an interdisciplinary project that focuses on the creation of a bilingual abstract corpus of bachelor’s theses from a linguistic point of view. The aim of this paper is to summarize the identified problem cases and to show the perspective of linguistic analysis of the corpus to be formed based on the research of special literature on the abstract types, the methodology of abstract analysis of bachelor’s theses and problems of text creation, as well as the results of the pilot study.From the annotated corpus of texts, 25 Latvian abstracts and their corresponding 25 English translations were randomly selected. The texts were obtained from the Registry of Final Theses of Riga Technical University (RTU) (2023–2024). When looking at the abstracts in correlation with the RTU methodological instructions, it can be established that the text type ‘abstract’ is mentioned and briefly described, but the descriptions and scope of their structure differ between faculties and study programs. In the pilot study, based on the modified models by Swales and Feak (2009) and Hyland (2000/2004) concerning the structure of abstracts consisting of different moves and steps, the abstracts’ text-internal sequence as well as quantitative indicators, such as detailed breakdowns of moves used, the length of an abstract, etc., were determined. Moreover, the relationship between moves forming patterns and the model of moves was adapted to the needs of researching texts written by Latvian students, and problems encountered during the intentional and deliberate annotation of the corpus were identified. These problems are mainly related to the lack of in-depth academic writing courses and the often-overgeneralized style of methodological instructions. The sequence of the moves and steps in the corpus is diverse. As abstracts constitute an internationally standardized text type, it does not seem purposeful to interpret differences in an intercultural context. This study has also found that the editing of abstract translations should be taught to students due to the way these translations are performed. The results of the pilot study show the need for modern academic writing support, which is the focus of further research