Co znaczy i co oznacza śmiech? O metonimicznych źródłach wieloznaczności czasownika śmiać się

Artykuł wpisuje się w nurt badań nad walencją i semantyką polskich czasowników nawiązujących do modelu struktur predykatowo-argumentowych autorstwa Stanisława Karolaka. Głównymi zadaniami pracy są repartycja i opis znaczeń polisemicznego kształtu językowego śmiać się. Zebrany materiał badawczy pokazuje, że predykat ten konstytuuje w polszczyźnie kilka różnych konstrukcji składniowych: [ktoś] śmieje się, że_, [ktoś] śmieje się z [kimś] z tego, że_, [ktoś] śmieje się do [kogoś], [ktoś] śmieje się z [siebie], że_, [ktoś] śmieje się z [kogoś/czyjegoś], że_, [ktoś] śmieje się z tego, że_, których struktury formalne, semantyczne i pojęciowe są rezultatem rozszerzeń metonimicznych opartych na mechanizmie polisemii radialno-łańcuchowej. Autorka podejmuje analizę procesów metonimicznych kształtujących gniazdo leksykalne utworzone wokół czasownika śmiać się oraz zwraca uwagę na kwestię, jak za ich sprawą do wtórnych znaczeń tego leksemu przenikają treści pochodzące ze sfery kultury. What does laughter mean from a linguistic and extralinguistic perspective? On metonymy as a source of polysemy of the Polish verb śmiać się (to laugh) Summary: This paper investigates the main syntactic and semantic features of the Polish polysemous predicate śmiać się (to laugh) within the theory of Semantic Syntax (Karolak 1984, 2002). The analysis shows that the verb forms several predicate-argument structures, e.g.: 1) [ktoś] śmieje się z [kimś] z tego, że_, ([somebody] laughs with [somebody] + SENT), 2) [ktoś] śmieje się do [kogoś] ([somebody] laughs to [somebody], which means ‘to grin’), 3) [ktoś] śmieje się z [siebie], że_ ([somebody] laughs at [oneself] + SENT), 4) [ktoś] śmieje się z [kogoś/czyjegoś], że_ ([somebody] laughs at [somebody / something that belongs to somebody] + SENT), which are different in terms of the number, status and semantic roles of their arguments, reference and type of humour incorporated in the act of laughter. The source of this variety of meanings of the predicate śmiać się (to laugh) lies in metonymic extensions which fit into the pattern of radial and chain polysemy. The article also outlines how the semiotic (cultural) meaning of laughter is reflected in the semantics of the verb śmiać się (to laugh) through the mechanism of metonymy.

From Tokenization to Self-Supervision: Building a High-Performance Information Extraction System for Chemical Reactions in Patents

Chemical reactions and experimental conditions are fundamental information for chemical research and pharmaceutical applications. However, the latest information of chemical reactions is usually embedded in the free text of patents. The rapidly accumulating chemical patents urge automatic tools based on natural language processing (NLP) techniques for efficient and accurate information extraction. This work describes the participation of the Melax Tech team in the CLEF 2020—ChEMU Task of Chemical Reaction Extraction from Patent. The task consisted of two subtasks: (1) named entity recognition to identify compounds and different semantic roles in the chemical reaction and (2) event extraction to identify event triggers of chemical reaction and their relations with the semantic roles recognized in subtask 1. To build an end-to-end system with high performance, multiple strategies tailored to chemical patents were applied and evaluated, ranging from optimizing the tokenization, pre-training patent language models based on self-supervision, to domain knowledge-based rules. Our hybrid approaches combining different strategies achieved state-of-the-art results in both subtasks, with the top-ranked F1 of 0.957 for entity recognition and the top-ranked F1 of 0.9536 for event extraction, indicating that the proposed approaches are promising.

Immediate recall as a secondary text: Referential parameters, pragmatics and propositions

Since the process of recalling combines comprehension and speech production, it is viewed as an extremely complex though understudied linguo-cognitive phenomenon. Recalls as secondary texts or text derivatives are also considered to be a good material to explore cognitive aspects of secondary texts production, information conversion procedures and types of transformations of primary texts. The notion of secondary texts also implies multiplicity, as an original text may be retranslated into numerous secondary texts different in quality and degree of completeness. The purpose of the study is to model the propositional secondary retold texts and to identify the specifics of the recipients interpretation of the main event in the text. It is aimed at discriminating the differences between the primary expository text and its 134 immediate recalls produced by 15-year old native Russian speakers. In order to reveal the specifics of the propositional content of a primary expository text and its recalls, their recipients used the following methodological operations: the description and interpretation of the semantic roles of the first and second arguments aligned to predicates on the basis of the verbs semantic properties; the employment of the psycholinguistic model of the utterances generation; the characteristic of memory as a complex of cognitive and mnemic processes; the definition of cognitive-semantic discourse structures; and the understanding of a proposition as a stable component of an utterance independent of the surface grammar. The comparison of the original text and its recalls with the use of innovative denotative maps enabled us to define successful and unsuccessful expression of propositional structures and the main idea of the original text. The classification of texts includes four groups based on the number of the reproduced propositions and types (weak or successful) of the reflection of the primary text denotative card. The authors designed and successfully implemented an innovative 11 stage-algorithm of revealing patterns of a printed text comprehension and its immediate recalls including the primary visual perception of the text, its primary interpretation, reading, encoding, reflection, preparation for an oral presentation, desobjectivation (distribution of semantic roles), interpretation, reflection, oral implementation and text. The work fills in certain gaps in the research, such as the specifics of immediate recalls production, identification of changes in propositional structures of immediate recalls, and expanding the corpus of semantic roles similar to Frame Net. The findings can be successfully applied in natural language processing and linguistic didactics.

Модели элементарных простых предложений со структурной схемой (S) + NAcc + VF в обобщенно-личных пословицах башкирского языка (на материале фольклорного корпуса)

Introduction. Modern philological science associates with structural-semantic approaches to the study of linguistic phenomena. But so far, the structural-semantic study of Bashkir proverbs based on the entire corpus, as well as the identification of the final types of models has not been undertaken. The purpose of this article is to study models of elementary simple sentences of the structural type (S) + NAcc + VF. This type, with an object in the accusative case, is the most frequent one; the accusative in the Bashkir language is a grammatical means of formalizing the actant as an object of direct immediate impact. Data and methods. Generalized personal proverbs selected from the folklore database of the Machine Corpus of the Bashkir language were used as the research material. The proverb corpus contains 354 proverbial expressions representing generalized personal sentences, i.e., 73.5 percent of the total number of one-part verb proverbs. The research method consisted in isolating an elementary simple sentence, which includes a predicate and obligatory actants; the author draws here on the work of Novosibirsk syntax students. The syntactic model is understood as the content plane of an elementary simple sentence, which is determined by its proposition. The structural diagram represents the expression plane of an elementary simple sentence. At this stage, the study focuses on proverbs with simple finite verbs. Results. The study shows that 8 models are implemented within the structural scheme (S) + NAcc + VF. Of these the model of physical impact on the object [(S) + NPatAcc + VActf] is the most frequent one, 62 proverbial expressions have been identified for this model. Each model of the structure in question is characterized by its inherent set of semantic roles and typical meaning, while in most cases, object actants act in the form of a certain accusative. Conclusions. The syntactic features characteristic of the construction of this type of proverbial expressions are likely to occur in modern literary language and colloquial speech with a more complex syntactic structure.

Dative radial category in Old Church Slavonic and Modern Russian

In this paper, the semantic roles expressed by the Dative case in Modern Russian and Old Church Slavonic are described in terms of radial categories. The corpus data shows that the radial category of the Dative case has changed since Old Church Slavonic. The radial category in Modern Russian is smaller, and it includes fewer subcategories than attested in Old Church Slavonic. The change of the category prototype could explain the changes in the category of the Dative case. Recipient is postulated to be the prototype of the Dative case category in Modern Russian, while Direction appeared to be the best prototype for Old Church Slavonic data.

The functions of the dative in the Jewish Arabic dialect of Baghdad

Like in Classical Arabic and other modern Arabic dialects, the preposition l- marks the dative also in the Jewish Arabic dialect of Baghdad (JB). Under the scope of the syntactic category of dative, one finds different semantic roles like recipients, benefactives, possessors, experiencers, and others. Moreover, some datives operate on the pragmatic rather than the semantic level of the clause. This paper defines and exemplifies seven different dative roles in JB based on their interpretive properties and accounts for their distinctive syntactic, semantic, and pragmatic features.

Polyadic quantification in hybrid coordination

The aim of this paper is to provide a syntactico-semantic analysis of hybrid coordination, in which what is coordinated are phrases bearing different grammatical functions and different semantic roles. The proposed account improves on previous HPSG analyses by giving up the assumption that all conjuncts are dependents of the same head and, more importantly, by taking into account the syntax–semantics interface and providing semantic representations. This aspect of the analysis builds on and generalizes previous HPSG work on polyadic quantification.

Deep Semantic Parsing with Upper Ontologies

This paper presents a new method for semantic parsing with upper ontologies using FrameNet annotations and BERT-based sentence context distributed representations. The proposed method leverages WordNet upper ontology mapping and PropBank-style semantic role labeling and it is designed for long text parsing. Given a PropBank, FrameNet and WordNet-labeled corpus, a model is proposed that annotates the set of semantic roles with upper ontology concept names. These annotations are used for the identification of predicates and arguments that are relevant for virtual reality simulators in a 3D world with a built-in physics engine. It is shown that state-of-the-art results can be achieved in relation to semantic role labeling with upper ontology concepts. Additionally, a manually annotated corpus was created using this new method and is presented in this study. It is suggested as a benchmark for future studies relevant to semantic parsing.

Application of Morphosyntactic Cues in Detection of GOAL Semantic Role

This paper analyzes the semantics of verbs with the prefix “do-” and explains the adlativity feature based on the morpho-syntactically annotated corpus hrWaC and handcrafted verb valency frames. The work aims to automatically add all types of adlativity to Croatian verb valency lexicon. As a result, it was revealed that if a language resource encodes “do-” as the adlative prefix in Croatian as a source language, then the adlative meaning in the target language can be assumed as well. Using the valency frame transition rules for language pairs, it is possible to design matching verb valency frames in other languages and consequently describe each verb and its translation by semantic roles (agent, patient, direction-to, and goal) and by selectional restrictions.