Text Categories Research Articles

The Sentiment Analysis Model of Services Providers’ Feedback

The purpose of this paper is to develop a hybrid model Ukrainian language sentiment analyzer, which should improve the accuracy of the mood definition to expand the Ukrainian language among the instruments on the market. The object of research is the processes of determining the language of the text and predicting its sentiment score. The subject of the study is Ukrainian comments posted by Google Maps users. The following text categories are taken into account: food, hotels, museums, and shops. The new method was built as an ensemble of support vector machine, logistic regression, and XGBoost, in combination with a rule-based algorithm. The practical use of the algorithm makes it possible to analyze the Ukrainian text in accordance with the category with the visualization of the research results. The accuracy of the proposed method is bigger than 0.88 in the worst case. The mining procedure of the positive and negative sides of service providers based on users’ feedback is developed. It allows electronics business to make improvements based on frequent positive and negative words.

An improved aspect-category sentiment analysis model for text sentiment analysis based on RoBERTa

The aspect-category sentiment analysis can provide more and deeper information than the document-level sentiment analysis, because it aims to predict the sentiment polarities of different aspect categories in the same text. The main challenge of aspect-category sentiment analysis is that different aspect categories may present different polarities in the same text. Previous studies combine the Long Short-Term Memory (LSTM) and attention mechanism to predict the sentiment polarity of the given aspect category, but the LSTM-based methods are not really bidirectional text feature extraction methods. In this paper, we propose a multi-task aspect-category sentiment analysis model based on RoBERTa (Robustly Optimized BERT Pre-training Approach). Treating each aspect category as a subtask, we employ the RoBERTa based on deep bidirectional Transformer to extract features from both text and aspect tokens, and apply the cross-attention mechanism to guide the model to focus on the features most relevant to the given aspect category. According to the experimental results, the proposed model outperforms other models for comparison in aspect-category sentiment analysis. Furthermore, the influencing factors of our proposed model are also analyzed.

DISCOURSE DIRECTION OF RESEARCH OF ARTISTIC TEXT IN MODERN LANGUAGE UKRAINIAN STUDIES

Background. One of the main issues of linguistics of the XX – early XXI century is a comprehensive study of the concept of discourse, which reflects a variety of real manifestations of human language communication in certain social spheres of its communicative space. The actualization of this concept in various scientific fields, the representatives of which study the essence and mechanisms of interaction of constants of such linguistic and philosophical dichotomies as “language and society”, “language and man”, has led to the need for its clear definition and typology of discourses etc. Despite the fact that linguists use a number of methods of discursive analysis of linguistic units, there is currently no single approach to their study within the theory of discourse.The purpose of the article is to identify and substantiate the features of the literary text in terms of modern discourse. The main objectives of the article: to determine the typological and categorical features of artistic discourse, to establish and describe the most important aspects of the study of artistic texts of different genres within discourse.Methods. The paper uses general scientific methods of observation, analysis, synthesis, comparison; linguistic methods, including the descriptive method – to interpret the discursive status of the literary text; the method of integrative analysis, which is the basis of a comprehensive study of a set of different aspects used in the study of the literary text as the embodiment of individualauthorial communicative intentions; the contextual-interpretive method – to identify probable reading strategies of reception of the literary text and its components.Results. The concept of discourse in the broadest sense is associated with all manifestations of communication. Discourse is a consequence of modeling knowledge structures that provide strategic planning, flow and control of communication, the operation of mechanisms of communicative competence, regulation of interactivity processes and more. An important communicative-semantic segment of the holistic nature of the analyzed concept is its interaction with various factors that reflect the essence of communication, in particular with the concept of the text. The linguistic tradition still has a great influence on the development of theoretical principles of categorization and typology of the concept of discourse, because until recently the discourse was often identified with the text as a complex linguistic sign that arises as a result of the process of language formation. Currently, researchers outline the structural and semantic parameters of the text with a number of features that also appear relevant and objective for the discourse: completeness; coherence; integrity; lexical, grammatical, logical, stylistic, semantic coherence; compositional completeness; communicative and pragmatic orientation, etc. Differentiation of discourses is one of the most important issues of modern discourse, which is currently a problem that researchers, based on a wide range of different socio-linguistic manifestations of communication, solve differently. To establish a typology of discourses, as well as texts, it is necessary to develop common criteria for their classification and determine the main categories of text and discourse. Artistic discourse, in contrast to other discourses, involves indirect communication of the author (sender) with numerous readers (addressees). When creating a text, the writer predicts the expected reaction to its content from readers, instead of “translating” the literary text into the discourse, the recipients carry out in their own way, based on their own knowledge, experience, feelings and emotions. The interaction of the addressee author with numerous addressees is devoid of real communicative factors, in particular temporal and local fixation. Despite the lack of direct contact, the writer's communicative influence on readers is obviously effective: if the artist has a deep knowledge of the language, individualizes and emphasizes linguistic means, it evokes a certain attitude to his literary texts and their evaluation by readers.Discussion. At the present stage of development of texts of any styles and genres requires discursive analysis, because in the language of fiction recorded authorial intentions that should “decipher” speakers in the process of dynamic cognition of semantic and semantic depths of a text, involving their own practical experience and knowledge during its decoding. We see prospects for further research of this topical problem in the study of features of artistic discourses of classic writers and famous contemporary artists, whose language creation is organically connected with the development of the Ukrainian literary word as an important component of the linguistic and cultural universe.

Category of tonality in the genre of essay (on the example of essays of V. M. Peskov)

The subject of this research is the category of tonality — one of the key text categories that reflects mentality of the author of the text. Essay, as a peculiar genre that functions at the intersection of publicistic and fiction styles, has its own typical range of tonality. Examination of the category of tonality in essays is a relevant  trend in the humanities. Well-known Soviet essays written by V. M. Peskov, which are considered exemplary, representative, and retain their relevance, served as the material for this work. The conducted analysis of V. M. Peskov’s essays reveals versatile means of explication of tonality and individual manner of the author. The novelty consists in observation that the texts of V. M. Peskov reflect the general tone of freedom and friendliness, the tone of conversation with the audience. His essays also reflect a clear authorial “Self”. Different essays mark different means of explication of tonality: in essays dedicated to nature, the author often applies metaphors, personifications, neutral words with positive connotations and words that express rational evaluation for imagery depiction of landscape with love for nature; in essays dedicated to people, the essayist often uses neutral lexicon with rational-evaluative or positive-evaluative connotations for creating image of the character; in essays dedicated to local culture, V. M. Peskov often applied emotionally expressive lexicon, words with suffixes of subjective evaluation, expressively syntactic constructs, special visual expressive means (personification, epithets, hyperbole, etc.) and colloquialism for admirable depiction of the local life and culture. Different techniques of explication of the category of tonality impose descriptiveness, attractiveness and influential power on the essay.

The category of chronotope in Akathist genre

The object of this research is the category of chronotope – a significant complex text category that reflects the semantic coordinates of the text and extra-textual reality. The category of chronotope intertwines the isomorphic components of textual time and textual space. The Akathist genre belongs to an insufficiently studied functional religious style. Examination of the category of chronotope in the context of Akathist genre is relevant trend in the modern functional stylistics. The research material contains the texts of Orthodox Akathists dedicated to the canonized saints. The objectivity of scientific research is justified by the choice of texts from various time periods. The author applies the method of categorical-textual analysis, which reveals the linguo-stylistic technique of formation of the components of general communicative-pragmatic meaning of the text. The novelty is defined by the fact that the use of categorical-textual method in the analysis of the Akathist type of text, interpreted as a generalized textual structure with typical genre-forming components, allows determining with a high degree of objectivity the genre specificity of Akathist and religious functional style. As a result of the conducted research, the author highlights the categorical-textual typology of chronotope, which characterizes the systematic uniformity of Akathist texts. The article describes the methods of explication of the sacred, objective (objectively-sacred and objectively-profane), and subjective (subjectively-sacred and subjectively-profane) types of time and space. Temporal indicators and locative units of the determined types of chronotope reconstruct the dichotomy of the divine and the earthly in the Akathist, which is inherent to the Christian worldview and constructively substantiates the uniqueness of the religious functional style. The conducted analysis reveals the communicative-pragmatic focus of the complex textual category of chronotope. The research materials can be used as an illustrative textbook in the practical course of the stylistics of modern Russian language.

Text categorization: past and present

Automatic text categorization is the operation of sorting out the text documents into pre-defined text categories using some machine learning algorithms. Normally, it defines the most important approaches to organizing and making the use of a large volume of information exists in unstructured form. Nowadays, text categorization is becoming an extensively researched field of text mining and processing of languages. Word sense, semantic relationships among terms, text documents and categories are quite essential in order of enhancing the performances of categorization. Various surveys on text categorization have already been available which involve techniques of various text representation schemes to such extent but do not include several approaches that have been explored in text categorization over the standard techniques. Here, an exhaustive analysis of different text categorization approaches over the conventional approaches has been undertaken. This survey paper explores a wide variety of algorithms used for categorizing text documents and tries to assemble the existing works into three basic fields: conventional methods, fuzzy logic-based methods, deep learning-based methods. Further, conventional methods have been categorized into three fields: text categorization using handcrafted features, text categorization using nature-inspired algorithms and text categorization using graph-based methods. Furthermore, this survey provides a clear idea about the available libraries used for different algorithms, availability of datasets, categorization technologies explored in various non-Indian and Indian languages as well.

Written Documents Analyzed as Nature-Inspired Processes: Persistence, Anti-Persistence, and Random Walks—We Remember, as Along Came Writing—T. Holopainen

Written communication is pivotal for societies to develop. However, lexicon and depth of information vary greatly among texts according to their purpose. Scientific texts, diffusion of science reports, general and area-specific news are all written differently. Thus, we explore the characterization of different text categories through a nature-inspired feature known as the Hurst parameter. We contend that the Hurst exponent is useful to unveil the rhetorical structure within written documents. We collected and processed texts in five categories: scientific articles, diffusion of science reports, business news, entertainment news, and random texts. Each category contains 350 documents. We found that the median for scientific texts has the highest value of the Hurst parameter (0.575), followed by business news (0.54); the median for randomly-generated texts is 0.48, which lies in the region associated with random walks. The median value for diffusion texts is 0.49, and for entertainment texts is 0.53. However, these two categories present high dispersion. We conclude that the Hurst parameter is a measure that quantifies the structure of communication in the selected categories of texts. Application of our finding in the field of e-research is discussed.

A corpus study of verbs in opinion articles of The Jakarta Post and the relation with text characteristics

As one of the text categories, opinion texts have distinctive characteristics compared to any other texts in newspapers, including the choice of verb usage. This study then aims at preliminarily examining the verbs used in opinion articles in The Jakarta Post to find out the relation between frequency and text characteristics. This study collected the opinion articles of The Jakarta Post comprising 47.143 words. This study was assisted by Lancsbox to store the corpus of opinion section texts, to identify the verb lemmas, and to count the frequency of verbs. The verbs found in this study were then classified based on Scheibman’s main verb classification (which is based on Halliday’s and Dixon’s verb types). The results of the study show that there are three most frequent verb types used in opinion texts in The Jakarta Post; they are material, verbal, and feeling verb types. Meanwhile, the lesser frequent ones are perception, possessive/relational, relational, and cognition verbs types. Meanwhile, the least frequent verb types are existential, corporeal and perception/relational verbs types. As opinion text conveys the argument of the writer, it is plausible to find feeling verb type belongs to the third most frequent types, along with material type to show concrete actions and verbal type to report the information. These frequencies exhibit that there is a firm relationship between text characteristics and the tendency of verb choice.

Theoretical Aspects of Scholarly Publishing about the Internet in Spanish Communication Journals

Theoretical aspects of scholarly publishing about the Internet in communication sciences in Spain have received little attention. The present text analyses scientific framework, categories, concepts and keywords used in research, collected from the most relevant specialized Spanish journals in the field, as well as research objectives that are pursued in connection to communication levels of study and types of data. A content analysis of a representative sample of 227 scientific articles was done in the five leading Spanish journals in communication in the period 1995–2015, in which the academic interesting on Internet as an object of study was consolidated. The results show a predominance of descriptive theoretical frameworks and a hegemony of journalism as an academic reference. Nevertheless, there is an increase complexity out of the mass media field. The research on the Internet in the communication field is presented as a reflexive opportunity to understand interdisciplinarity and the way this acquires epistemological consistence in the scientific discourse.

Typographic Cueing Facilitates Survey Completion on Smartphones in Older Adults

As more people use mobile devices to complete Web surveys, it is critical to verify that current Web survey design practices carry over to mobile Web. One such practice is to bold the question to draw attention to the start of each question and to italicize instructions to facilitate filtering of potentially unneeded information. We conducted a laboratory experiment to investigate whether the common Web survey design practice of utilizing typographic cues to indicate different categories of question text is beneficial for surveys designed for smartphones. Thirty participants completed a 5-question survey on an iPhone. Each participant was randomly assigned to one of the four designs: (1) bolded questions and italicized instructions, (2) bolded questions and plain instructions, (3) plain questions and italicized instructions, (4) plain questions and plain instructions. Survey performance was measured with survey completion time, difficulty rating, and preference. We found that italicizing the instructions led to shorter completion times, while bolded questions had no effect on performance. All designs were rated easy to use. The designs with bolded questions were preferred by 83% of the participants. We argue that italicized instructions assist in visual filtering by providing a cue to respondents that they may not need to read the instructions in detail but may scan or skim the instructions instead. Scanning or skimming leads to shorter completion time without detriment to accuracy because respondents can still acquire the relevant instructions.

A Topical Category-Aware Neural Text Summarizer

The advent of the sequence-to-sequence model and the attention mechanism has increased the comprehension and readability of automatically generated summaries. However, most previous studies on text summarization have focused on generating or extracting sentences only from an original text, even though every text has a latent topic category. That is, even if a topic category helps improve the summarization quality, there have been no efforts to utilize such information in text summarization. Therefore, this paper proposes a novel topical category-aware neural text summarizer which is differentiated from legacy neural summarizers in that it reflects the topic category of an original text into generating a summary. The proposed summarizer adopts the class activation map (CAM) as topical influence of the words in the original text. Since the CAM excerpts the words relevant to a specific category from the text, it allows the attention mechanism to be influenced by the topic category. As a result, the proposed neural summarizer reflects the topical information of a text as well as the content information into a summary by combining the attention mechanism and CAM. The experiments on The New York Times Annotated Corpus show that the proposed model outperforms the legacy attention-based sequence-to-sequence model, which proves that it is effective at reflecting a topic category into automatic summarization.

Toward neuroadaptive support technologies for improving digital reading: a passive BCI-based assessment of mental workload imposed by text difficulty and presentation speed during reading

We investigated whether a passive brain–computer interface that was trained to distinguish low and high mental workload in the electroencephalogram (EEG) can be used to identify (1) texts of different readability difficulties and (2) texts read at different presentation speeds. For twelve subjects we calibrated a subject-dependent, but task-independent predictive model classifying mental workload. We then recorded EEG data from each subject, while twelve texts in blocks of three were presented to them word by word. Half of the texts were easy, and the other half were difficult texts according to classic reading formulas. From each text category three texts were read at a self-adjusted comfortable presentation speed and the other three at an increased speed. For each subject we applied the predictive model to EEG data of each word of the twelve texts. We found that the resulting predictive values for mental workload were higher for difficult texts than for easy texts. Predictive values from texts presented at an increased speed were also higher than for those presented at a normal self-adjusted speed. The results suggest that the task-independent predictive model can be used on single-subject level to build a highly predictive user model of the reader over time. Such a model could be employed in a system which continuously monitors brain activity related to mental workload and adapts to specific reader’s abilities and characteristics by adjusting the difficulty of text materials and the way it is presented to the reader in real time. A neuroadaptive system like this could foster efficient reading and text-based learning by keeping readers’ mental workload levels at an individually optimal level.

Category of theme in the genre of essay (on the materials of V. M. Peskov’s essays)

The subject of this research is the category of theme – one of the leading text categories that reflects the subject matter of speech and extra-textual reality. Analysis of the category of theme in the essay genre is a relevant trend in modern linguistics. The essays by V. M. Peskov, whose works are considered exemplary and representative for the scientific research in the field of humanities, serve as the materials for this article. Among the essays of the Soviet period, the essays of V. M. Peskov rightfully hold a special place and retain its relevance. The research methodology contains categorical-textual analysis based on the category of communicativeness. The novelty consists in application of the categorical-textual method towards analyzing the genre of essay, which implies a combination of uniform by means of explication textual categories, and with a high degree of objectivity, allows to determine the characterological features of the genre and functional style. The means of explication of the category of theme in the genre of essay have the following peculiarities: incorporation of several functions in a single nomination (primary, basic, principal); application of diverse additional nominations (referentially identical nominations, synonyms, taxonomic nominations, substitutions, grammatical transformations, zero nominations). The analysis of a wide range of additional nominations demonstrates that the essays by V. M. Peskov with a high degree of confidence can be attributed to the field of publicistic, fiction and colloquial styles; or in other words, the genre is at the intersection of several functional styles.

Многообразие и полифония текстов в музыкальной культуре эпохи романтизма

The article studies the phenomenon of a musical text, identifies the features distinguishing it from a piece of work. The core aspects of creativity of renowned Romantic composers of Western Europe and Russia in the XIX - early XX century are specified. The work also analyses the notion of text polyphony, peculiarities of its manifestation in the Western European and Russian musical art of the Romantic period. The research aims to provide a description of what the text category is in musical culture of the Romantic period. The study is novel in that the phenomenon of a text in musical culture is understudied and thematic variety of musical works is a “calling card” of the Romantic period. Basing on the research findings, the author sums up specifics of variety of musical texts in musical culture of the Romantic period.

METACOMMUNICATIVE REFLECTION IN THE "NAIVE" EPISTOLARY

The study of private correspondence language has been a relevant problem in Ukrainian philology for a long time, however linguists are primarily interested in the epistolary works of elitist language personalities, including Ukrainian writers, cultural and public figures, and therefore analyze the codified Ukrainian speaking. The paper suggests changes to the scientific vector, directing it to the written communication of "naive" authors and accordingly to the Ukrainian non-codified spoken language in its written form. Our research aim is to analyze the reflexives of "naive" authors of daily family and household letters of epistolary communication as a whole and its individual elements. In other words, this work examines those linguistic units (metacommunicatives) that transmit information about the letters, the correspondence process in the letters. The tasks of the work are to identify and describe the metacommunicative reflexives as the subject of research, as well as to establish their functional load in the "body" of the epistolary text. The research material was a collected corpus of daily family and household letters, whose authors are Ukrainians as native speakers, rarely – Russian, living in the Sumy Oblast, and used surzhyk in their daily communication. We used the descriptive, lexical-semantic and discursive analysis methods to establish the composition of the metacommunicative reflexives, which include neutral and connotative marked names of letters and other forms of distance communication (postcard, telegram, parcel, money transfer, phone call), the most common definitions and marking the procedurality of epistolary communication. In addition, the names of material carrier of the letters, tool names for writing the letter, the verbal feature of the individual manner of writing, etc. were introduced into the circle of metacommunities. The letter composition, implementation of text categories (integrities and coherences in priority), construction of epistolary chronotype, expression of emotions, and feelings are among the functions performed by the metacommunicative reflexives.

Функциональный потенциал интертекстуальности в современных англоязычных сказках

The study aims to determine functional capabilities of the intertextuality text category using the material of the modern English-language fairy-tales. Autointertextuality is mentioned and examples of this phenomenon present in the author’s fairy-tale texts are given. Scientific novelty of the research lies in collecting new data on functioning of the intertextuality category and describing new patterns in using functionality of the intertextuality category by the authors. It has been found that the use of intertextual relations in the modern English-language fairy-tales contributes to realisation of such author’s intentions as creating a comic effect, adding factuality to a narrative, organising a chronotope, developing a plot, characterisation and setting.

THE STATUS OF CAPITAL IN THE “KYIV’S TEXT” OF THE BAROQUE EPOCH

Urban studies are a multidisciplinary area, but experts in different fields conclude that the city is worth considering in the categories of text. Moreover, urban studies in literary criticism are distinguished by the fact that the phrase “a city as a text” for philologists is not only a beautiful metaphor. The fiction space, including urban space, is a separate reality that exists according to its own rules, depending on the epoch style and the genre of a particular piece of writing. In the Baroque times, a city is a place of creation and functioning of culture. In the XVII century in Ukrainian cities appeared many educational institutions, thanks to patrons, numerous churches were built, which was reflected in panegyrics. At the same time, the large number of Kyiv Rus’ buildings had been reconstructed, and there are mentions of the Russ past of the city in polemical literature, school dramaturgy and chronicles. In Kyiv, the glory of the “capital”, “Jerusalem of the Russ land” is affirmed. The purpose of the author is to explore the mechanism of rethinking the past and its role in the construction of an artistic model of the city in the Baroque epoch. The ways of transcoding Kyiv Russ urban motifs into the language of Baroque culture are considered in the paper. In the early 1600s, Kyiv remained a capital status in the minds of citizens, despite the decline and destruction of the past. At the turn of the 16–17th centuries the idea of continuity of the history and glory of Kyiv from the Middle ages became widespread among intellectuals. In times of statelessness, the current becomes relative and unimportant, while the past is considered to be the actual reality. This article aims to examine how the urban space becomes the embodiment of collective memory: through the buildings of St. Sophia’s Cathedral, St. Michael’s Cathedral, the Desiatynna Church, the Church of Virgin Mary Pyrohoshcha, the Cathedral of the Dormition in the Kyiv-Pechersk Monastery, the holiness and centrality of the city are transmitted in sacral and profane levels. The author pays particular attention to the analysis of the opposite self-image of Kyiv citizens as residents of the “city on the outskirt”, on the border of Wild Field.

Глаголы, актуализирующие семантику модальности сомнения, в текстах англоязычного публицистического дискурса

The article examines specifics of actualising doubt modality by verbs in texts of publicistic discourse. Doubt modality is presented as a text category with the general meaning of uncertainty. The study showed that actualisation of this meaning occurs under the influence of context, within the boundaries of a text as a whole maintained by interaction of means belonging to different linguistic levels. And the verbs whose semantic structure registers the meaning of uncertainty as a secondary meaning or non-existent one are able to exhibit doubt modality acting as direct and indirect means of expressing this semantics. The study is novel in that it ascertains dependency of explicit or implicit expression of doubt modality by verbs on syntactic structures of corresponding utterances.

Joint Character-Level Convolutional and Generative Adversarial Networks for Text Classification

With the continuous renewal of text classification rules, text classifiers need more powerful generalization ability to process the datasets with new text categories or small training samples. In this paper, we propose a text classification framework under insufficient training sample conditions. In the framework, we first quantify the texts by a character-level convolutional neural network and input the textual features into an adversarial network and a classifier, respectively. Then, we use the real textual features to train a generator and a discriminator so as to make the distribution of generated data consistent with that of real data. Finally, the classifier is cooperatively trained by real data and generated data. Extensive experimental validation on four public datasets demonstrates that our method significantly performs better than the comparative methods.

CatGAN: Category-Aware Generative Adversarial Networks with Hierarchical Evolutionary Learning for Category Text Generation

Generating multiple categories of texts is a challenging task and draws more and more attention. Since generative adversarial nets (GANs) have shown competitive results on general text generation, they are extended for category text generation in some previous works. However, the complicated model structures and learning strategies limit their performance and exacerbate the training instability. This paper proposes a category-aware GAN (CatGAN) which consists of an efficient category-aware model for category text generation and a hierarchical evolutionary learning algorithm for training our model. The category-aware model directly measures the gap between real samples and generated samples on each category, then reducing this gap will guide the model to generate high-quality category samples. The Gumbel-Softmax relaxation further frees our model from complicated learning strategies for updating CatGAN on discrete data. Moreover, only focusing on the sample quality normally leads the mode collapse problem, thus a hierarchical evolutionary learning algorithm is introduced to stabilize the training procedure and obtain the trade-off between quality and diversity while training CatGAN. Experimental results demonstrate that CatGAN outperforms most of the existing state-of-the-art methods.