Internet News Headlines: Ontological and Orthological Aspects

The author presents the results of a multidimensional analysis of Internet news headlines based on the headings of the Yandex news aggregator. The issue of the text status of news headlines is considered. When solving this problem, special attention is paid to the formation of correlative paradigms of headings, united by a common denotative meaning. Methods of semantic interaction of heading paradigms based on different types of topic-rhematic deployment are described. It is proved that the paradigms of headings, complementing each other in informational and pragmatic aspects, form the discourse of Internet headings. It is concluded that this way of functioning of headlines enhances their semantic and visual autonomy from the news text, which allows us to consider Internet news headlines as minitext. The frequency methods of lexico-syntactic transformations of the original headings are analyzed, on the basis of which the constituents of paradigms are formed. At the same time, attention is focused on the orthological aspect of Internet headers. The author comes to the conclusion that the focus on the variability and efficiency of headings leads not only to the appearance of lexical and grammatical errors, but also to their replication and consolidation in the mind of the addressee as a result of changes in the structure of cognitive models.

Knowledge-Enhanced Graph Attention Network for Fact Verification

Fact verification aims to evaluate the authenticity of a given claim based on the evidence sentences retrieved from Wikipedia articles. Existing works mainly leverage the natural language inference methods to model the semantic interaction of claim and evidence, or further employ the graph structure to capture the relation features between multiple evidences. However, previous methods have limited representation ability in encoding complicated units of claim and evidences, and thus cannot support sophisticated reasoning. In addition, a limited amount of supervisory signals lead to the graph encoder could not distinguish the distinctions of different graph structures and weaken the encoding ability. To address the above issues, we propose a Knowledge-Enhanced Graph Attention network (KEGA) for fact verification, which introduces a knowledge integration module to enhance the representation of claims and evidences by incorporating external knowledge. Moreover, KEGA leverages an auxiliary loss based on contrastive learning to fine-tune the graph attention encoder and learn the discriminative features for the evidence graph. Comprehensive experiments conducted on FEVER, a large-scale benchmark dataset for fact verification, demonstrate the superiority of our proposal in both the multi-evidences and single-evidence scenarios. In addition, our findings show that the background knowledge for words can effectively improve the model performance.

Full-Abstract Biomedical Relation Extraction with Keyword-Attentive Domain Knowledge Infusion

Relation extraction (RE) is an essential task in natural language processing. Given a context, RE aims to classify an entity-mention pair into a set of pre-defined relations. In the biomedical field, building an efficient and accurate RE system is critical for the construction of a domain knowledge base to support upper-level applications. Recent advances have witnessed a focus shift from sentence to document-level RE problems, which are more challenging due to the need for inter- and intra-sentence semantic reasoning. This type of distant dependency is difficult to understand and capture for a learning algorithm. To address the challenge, prior efforts either attempted to improve the cross sentence text representation or infuse domain or local knowledge into the model. Both strategies demonstrated efficacy on various datasets. In this paper, a keyword-attentive knowledge infusion strategy is proposed and integrated into BioBERT. A domain keyword collection mechanism is developed to discover the most relation-suggestive word tokens for bio-entities in a given context. By manipulating the attention masks, the model can be guided to focus on the semantic interaction between bio-entities linked by the keywords. We validated the proposed method on the Biocreative V Chemical Disease Relation dataset with an F1 of 75.6%, outperforming the state-of-the-art by 5.6%.

Semantic interoperability of interaction of elements in network-centric systems.

In the paper is proposed an approach for semantic interaction modelling of network-centric system elements and their context parameters based on the multi-agent approach and the systems, capabilities, operations, programs, and enterprises model for interoperability assessment. The peculiarity of this approach is that the elements of a network-centric system are formalized as agents of various types (human agents and technical cognitive agents), where the semantic interaction is determined by the agent goals, the subject area of interaction and the context. In addition, correct of interaction interpretation is determined the agent's knowledge model. The current study takes place as a part of Russian Foundation for basic research finance project no. 19-07-00774.

Hierarchical semantic interaction-based deep hashing network for cross-modal retrieval

Due to the high efficiency of hashing technology and the high abstraction of deep networks, deep hashing has achieved appealing effectiveness and efficiency for large-scale cross-modal retrieval. However, how to efficiently measure the similarity of fine-grained multi-labels for multi-modal data and thoroughly explore the intermediate layers specific information of networks are still two challenges for high-performance cross-modal hashing retrieval. Thus, in this paper, we propose a novel Hierarchical Semantic Interaction-based Deep Hashing Network (HSIDHN) for large-scale cross-modal retrieval. In the proposed HSIDHN, the multi-scale and fusion operations are first applied to each layer of the network. A Bidirectional Bi-linear Interaction (BBI) policy is then designed to achieve the hierarchical semantic interaction among different layers, such that the capability of hash representations can be enhanced. Moreover, a dual-similarity measurement (“hard” similarity and “soft” similarity) is designed to calculate the semantic similarity of different modality data, aiming to better preserve the semantic correlation of multi-labels. Extensive experiment results on two large-scale public datasets have shown that the performance of our HSIDHN is competitive to state-of-the-art deep cross-modal hashing methods.

EKPHRASIS AND META-EKPHRASIS IN JULIAN BARNES’S ESSAY “GÉRICAULT: CATASTROPHE INTO ART”: A COGNITIVE POETIC ANALYSIS

The article discusses the semantic aspects of ekphrasis and meta-ekphrasis in Julian Barnes’s essay “Géricault: Catastrophe into Art” from the cognitive poetics perspective. In his essay, Barnes dwells upon the history and interpretations of Géricault’s masterpiece which represents the survivors of the wreck of a French frigate in 1816. The aim of the study is to reveal the semantic integration of ekphrastic contexts (those parts of the essay which provide description of a painting) and meta-ekphrastic contexts (those parts of the essay which are not descriptions of a painting per se, however they only develop their meaning in connection with ekphrastic contexts). The article suggests using the term meta-ekphrasis to account for the textual contexts which while being semantically related to ekphrasis, do not offer a painting description but a narration about some events related with the painting or ideas inspired by looking at the painting. Used in this meaning, the term meta-ekphrasis is utilised in the paper to reveal the development of Barnes’s original idea about tragedy and art in his essay. The research is grounded in cognitive poetic approach to ekphrasis and employs cognitive poetic instruments of analysis. It presents the results which demonstrate that the main cognitive poetic means that ensure semantic interaction of ekphrastic and meta-ekphrastic contexts in Barnes’s essay are the following: dialogism, hypothetical modality and conceptual metaphors with the source domain of SEA NAVIGATION. It is the semantic integrity of ekphrastic and meta-ekphrastic contexts in Barnes’s essay that allows the writer to present his unconventional treatment of tragedy as being purposeful since it produces art. The article can be of interest to the scholars of sematic interaction between verbal and visual texts as well as cognitive poetic facets of prose texts.