Learning Sentence-to-Hashtags Semantic Mapping for Hashtag Recommendation on Microblogs

The growing use of microblogging platforms is generating a huge amount of posts that need effective methods to be classified and searched. In Twitter and other social media platforms, hashtags are exploited by users to facilitate the search, categorization, and spread of posts. Choosing the appropriate hashtags for a post is not always easy for users, and therefore posts are often published without hashtags or with hashtags not well defined. To deal with this issue, we propose a new model, called HASHET ( HAshtag recommendation using Sentence-to-Hashtag Embedding Translation ), aimed at suggesting a relevant set of hashtags for a given post. HASHET is based on two independent latent spaces for embedding the text of a post and the hashtags it contains. A mapping process based on a multi-layer perceptron is then used for learning a translation from the semantic features of the text to the latent representation of its hashtags. We evaluated the effectiveness of two language representation models for sentence embedding and tested different search strategies for semantic expansion, finding out that the combined use of BERT ( Bidirectional Encoder Representation from Transformer ) and a global expansion strategy leads to the best recommendation results. HASHET has been evaluated on two real-world case studies related to the 2016 United States presidential election and COVID-19 pandemic. The results reveal the effectiveness of HASHET in predicting one or more correct hashtags, with an average F -score up to 0.82 and a recommendation hit-rate up to 0.92. Our approach has been compared to the most relevant techniques used in the literature ( generative models , unsupervised models, and attention-based supervised models ) by achieving up to 15% improvement in F -score for the hashtag recommendation task and 9% for the topic discovery task.

The Verb שָׁאַף in Biblical Hebrew

Biblical Hebrew lexicons unanimously present the basic meaning of the verb שׁאף as “pant, snuff.” Absent etymological evidence, however, the lexical value of the verb hangs on the contextual interpretation of three attestations where the verb has not undergone semantic expansion: Isa 42:14; Jer 2:24; 14:6. Fresh analysis of the philological evidence garners support for an alternate interpretation of שׁאף רוח in Jer 2:24; 14:6 as “bray, cry out” and suggests that ואשׁאף in Isa 42:14 constitutes an elliptical form of the phrase with the same meaning. This new semantic understanding in turn allows for a reanalysis of derived meanings, furnishing a revised understanding of the verb שׁאף.

Etymology of the Word and Axiological-Evaluative Semantics

Etymology is a section of linguistics, a section of comparative-historical linguistics that studies the origin of language words; a set of research methods aimed at clarifying the origin of the word, as well as the result of this clarification. The etymology of the word is its linguistic and cultural-historical passport, its biography, which reflects the structural and semantic status of the word in the ancient period of language development and its place in the circle of related and unrelated languages. Etymology is designed to explain all the changes (or invariance) of the form of the word, at the same time all the metamorphoses of its meaning in the light of the cultural and historical conditions in which a language developed; rationally assess the potential inherent in the word for further development and trace the mechanisms of its semantic development. . The purpose of research is to clarify the role of etymology and axiology in semantic and etymological analysis. Semantic and etymological analysis of reference values as information centers of the grid of linguistic meanings of the nominative system reveals the original forms and original semantic dominants in the typological context, determine their potential for further semantic and nominative development in the languages studied. It is a kind of "bridge" to the new semantic links, correlated with those already established. Typological comparison allows to identify semasiological parallels (similar semantic dominants and similar semantic expansion) against the background of areal and genetic. Semasiological parallels are research-motivated, because we are dealing with one concept, which is expressed in languages with semasiologically common roots. Based on the analysis of the components of the onomasiological paradigm, we can distinguish two types of assessment: axiological evaluation on an objective basis and axiological evaluation on a subjective basis. Axiological evaluation on an objective basis focuses on rational evaluation, which researchers traditionally associate with the notion of stereotype used in logical evaluation theories. Axiological evaluation on a subjective basis. In this type of evaluation, the leading role belongs to the emotional component: the already evaluated phenomenon here is layered with the actual subjective evaluation, so the already evaluated objective feature is evaluated again, subjectively. It is a mix of a sign and its evaluation, another assessment is subjective, emotional. The formation of axiologically evaluative semantics of units of all types in the totality of meaning and form can be carried out on the basis of typological and specific for each language psychological associations of figurative and non- figurative nature, due to the presence of different people, different ideas and concepts. Language is an expression not only of the linguistic thinking of an individual nation, but also of the linguistic and cultural experience inherent in humanity as a whole.

Semantic expansion of search queries

One of the major challenges in Web search pertains to the correct interpretation of users' intent. Query Expansion is one of the well-known approaches for determining the intent of the user by addressing the vocabulary mismatch problem. A limitation of the current query expansion approaches is that the relations between the query words and the expanded words is limited. In this thesis, we capture users' intent through query expansion. We build on earlier work in the area by adopting a pseudo-relevance feedback approach; however, we advance the state of the art by proposing an approach for feature learning within the process of query expansion. In our work, we specifically consider the Wikipedia corpus as the feedback collection space and identify the best features within this context for term selection in two supervised and unsupervised models. We compare our work with state of the art query expansion techniques, the results of which show promising robustness and improved precision.

Semantic expansion of search queries

One of the major challenges in Web search pertains to the correct interpretation of users' intent. Query Expansion is one of the well-known approaches for determining the intent of the user by addressing the vocabulary mismatch problem. A limitation of the current query expansion approaches is that the relations between the query words and the expanded words is limited. In this thesis, we capture users' intent through query expansion. We build on earlier work in the area by adopting a pseudo-relevance feedback approach; however, we advance the state of the art by proposing an approach for feature learning within the process of query expansion. In our work, we specifically consider the Wikipedia corpus as the feedback collection space and identify the best features within this context for term selection in two supervised and unsupervised models. We compare our work with state of the art query expansion techniques, the results of which show promising robustness and improved precision.

A comparative study on the formation of Sinoxenic person nouns in Korean, Chinese, Japanese, and Vietnamese

This study attempts to sketch out how Korean, Chinese, Japanese, and Vietnamese (hereafter K, C, J, and V, respectively) have produced Sinoxenic personal nouns. The formation of Sinoxenic personal noun falls into one of three types: (1) with an affix attached to the original word; (2) semantic expansion; (3) expansion to figurative meanings. One observes all three types throughout KCJV; however, specific patterns differ slightly between Vietnamese and the other three languages. The differences between Vietnamese and the other three in Sinoxenic formation result from differences in geographical distances and frequencies of linguistic exchanges. Meanings of Sinoxenic words stayed similar among Korea, China, and Japan, due to frequent human and literary exchanges, which did not occur between Vietnam and the other three countries.

Pushkin’s allusions in the comedy of A. P. Chekhov "The Cherry Orchard"

The study of Pushkin's allusions in the works of Anton Chekhov has been carried out for several decades. The relevance and novelty of this topic is associated with new observations and searches for intertextual connections, the expansion of the context of analysis, including in connection with the appeal to the dramatic works of the writer. This work considers the allusions from the works of A. Pushkin "The Queen of Spades", "Boris Godunov" and others, their role in the artistic whole of the comedy of A. P. Chekhov "The Cherry Orchard". The effectiveness of the approach which was outlined in the studies of R.G. Nazirov in the 1990s, as well as S.A. Kibalnik, M.V. Litovchenko and other domestic scientists is confirmed. Literary allusions resonate in Chekhov's artistic world, cumulatively marking a transitional stage of cultural and historical life and its personal comprehension. The intertextuality of Chekhov's works allows us to create an integral image of Russian culture: the semantic relations of allusions and reminiscences are combined with the multilayered symbolism of sound, hearing, gesture, touch - in general, the experience of time and the multiple meanings of words. The semantic "expansion" of Chekhov's play text leads not only to comparisons with Russian classics, but also to an open finale - a future literary context, implementing the "principle of plot uncertainty" and removing the uniqueness/truthfulness of any interpretation.

The intersemiotic affordances of photography and poetry

This article explores the intersemiotic affordances of photography and poetry and the expansion of meaning that surpasses the meanings embedded in and elicited from both. We specifically investigate the processes and mechanisms of this semantic expansion by systematically reconstructing the compositional process of poems written from three photographs and forensically investigate how the poems emerged out of each visual frame. We discovered that intersemiosis between photography and poetry demonstrates a strong interpretative component. Intra-semiotic connections between elements within the photograph are interpreted by the viewer or writer and are translated by means of inter-semiotic triggers into intra-semiotic connections within the emerging poem during the process of composition. The resulting inter-semiotic connections between the photograph and the poem create and multiply meaning for both mediums together and independently. In other words, in the process of composition, the poem reads the meanings of components of the photograph framed by the photographer and super-frames them; creating a new frame of meanings that draw upon, and extend, meanings in the original frame of the photograph. At the same time, the poem enters a stage of self-change and self-reflection, inhabiting the life of the photograph.