Learning a Hierarchical Intent Model for Next-Item Recommendation

A session-based recommender system (SBRS) captures users’ evolving behaviors and recommends the next item by profiling users in terms of items in a session. User intent and user preference are two factors affecting his (her) decisions. Specifically, the former narrows the selection scope to some item types, while the latter helps to compare items of the same type. Most SBRSs assume one arbitrary user intent dominates a session when making a recommendation. However, this oversimplifies the reality that a session may involve multiple types of items conforming to different intents. In current SBRSs, items conforming to different user intents have cross-interference in profiling users for whom only one user intent is considered. Explicitly identifying and differentiating items conforming to various user intents can address this issue and model rich contextual information of a session. To this end, we design a framework modeling user intent and preference explicitly, which empowers the two factors to play their distinctive roles. Accordingly, we propose a key-array memory network (KA-MemNN) with a hierarchical intent tree to model coarse-to-fine user intents. The two-layer weighting unit (TLWU) in KA-MemNN detects user intents and generates intent-specific user profiles. Furthermore, the hierarchical semantic component (HSC) integrates multiple sets of intent-specific user profiles along with different user intent distributions to model a multi-intent user profile. The experimental results on real-world datasets demonstrate the superiority of KA-MemNN over selected state-of-the-art methods.

On Agree without agreement as a source of reflexive voice constructions

Switch-reference has recently been argued to be the result of clausal functional heads entering into Agree with two nearby noun phrases, creating pointers to those noun phrases but not actually copying their morphosyntactic features. Instead, the semantic component interprets the pointers as referential dependency holding between pointed-to noun phrases. This article applies this analysis to reflexive voice constructions in which a feature-invariant affix appears on the verb to indicate that the (highest, direct) object is referentially dependent on the (thematic) subject of the same clause. First it surveys the properties that such constructions should have if reflexive voice is maximally like switch-reference. Then it argues that the Bantu language Lubukusu has just such a construction, the verbal affix i partnering with the overt anaphor omweene to create reflexive clauses. Dravidian reflexive voices are presented as another possible case. Finally, it turns to reflexive and reciprocal voice constructions in Shipibo (Panoan), which seem to have a detransitivizing effect. However, no major detransitivizing account fits all the facts. Rather, reflexive voice in Shipibo is like Lubukusu, except that the anaphor is phonologically null and deficient in phi-features, failing to trigger ergative case on the subject for that reason. True detransitivization may happen in some languages with reflexive voice, but not in all, and it will take considerable care to sort out which are which.

Парные глаголы с каузативным значением: семантические особенности (на материале бурятского языка)

Introduction. The article discusses the semantics of paired causative verbs in the Buryat language, which has not been the subject of a special study yet. The aim was to study the semantic features of Buryat paired causative verbs by way of identifying: 1) the types of paired causative verbs, 2) the contribution of each semantic component of a verbt of its general semantics and their correlation, and 3) paired causative verbs as a special case in expressing causative semantics. Materials and methods. The data was collected from works of fiction in the Electronic Corpus of the Buryat language; contextual and distributive analysis were used as the primary methods of research. Results. Semantic analysis of paired causative verbs, illustrative of causative relationships in the linguistic picture of the Buryat world, indicated that in pairs of non-causative + causative verb the latter is a leading component, which demonstrates the power of causative semantics; the verbs of this type most often express various emotions associated with impact and subjective assessment. The other type are represented by pairs of synonymous causative verb + causative verb that are effective in expressing the intensity of the impact; with the semes of the synonymous verbs combined, the meaning of causation in such pairs is enhanced, and their expressiveness and emotionale valuation aspects come to the fore.

A content spectral-based text representation

In the last few years, text analysis has grown as a keystone in several domains for solving many real-world problems, such as machine translation, spam detection, and question answering, to mention a few. Many of these tasks can be approached by means of machine learning algorithms. Most of these algorithms take as input a transformation of the text in the form of feature vectors containing an abstraction of the content. Most of recent vector representations focus on the semantic component of text, however, we consider that also taking into account the lexical and syntactic components the abstraction of content could be beneficial for learning tasks. In this work, we propose a content spectral-based text representation applicable to machine learning algorithms for text analysis. This representation integrates the spectra from the lexical, syntactic, and semantic components of text producing an abstract image, which can also be treated by both, text and image learning algorithms. These components came from feature vectors of text. For demonstrating the goodness of our proposal, this was tested on text classification and complexity reading score prediction tasks obtaining promising results.

The Influence of Western Pop Culture on the Creation of the Image of the Performer Through the Design of Advertising Posters and Plates in the Period of Late Socialism

The aim of the research is to analyze the formation of the artist’s image through the design of musical posters and records in the period of late socialism by borrowing the semantic component and formal techniques in foreign pop culture, including graphic design, which served as an important component. The research methodology is based on a logical-analytical tool of scientific knowledge. Scientific novelty. Previous research has been supplemented by the analysis of new empirical data, including the design of record covers, visual coincidences of primary sources and borrowings. The revealed facts will be useful for the further development of both modern Ukrainian design and the formation of the artist’s image in show business. Conclusions. As a result of the study we conclude that in the period 1970-1990 in the process of formation of pop culture in the USSR the problem of its visual representation was formed, as evidenced by numerous pieces borrowed from foreign pop design solutions for advertising products: posters, placards, album covers, accompanying products. At an early stage, this trend was manifested despite the existence of the “Iron Curtain”, ideological isolation and was episodic. In the period chronologically close to the collapse of the USSR, such processes of stylistic citation in design became more frequent in line with the growth of information coming from the developed West. Copying Western techniques and visual language can be seen as a desire to break away from the dogmatic patterns of culture instilled by socialist realism, regardless of the conscious or unconscious nature of imitation. Blind imitation was later replaced by a reasonable rethinking, so in the best examples of domestic design of the period under study there were attempts to give formal Western methods the character of the local national color, which requires further exploration.

The image of Russia in the consciousness of Russian youth: peculiarities of the value-semantic code of cultural representations

The study involved 98 representatives of Moscow youth; 116 of Perm youth and 104 of Tyumen youth. A total 318 respondents, 44% young men and 56% young girls, aged 18 to 25. The 49 bipolar psychosemantic scale of the study describes the image of the country in terms of its strength and authority in the international arena, activity in transformations and various aspects of assessment. Respondents evaluated the images of “Russia-country”, “Future Russia”, “USA”, “China”. The categorical structures of young people's social representations about the country's image, identified in all three regions, have three substantively comparable factors: 1) “Welfare, progressiveness of the country”, 2) “Level of social distance”, 3) “Civilizational attribution”. The value-semantic component in the structure of young people's ideas fixes the cultural codes of Russian civilization: “beauty”, “generosity”, “mercy”, “kindness”, “spirituality”, “morality”.

Meaning in LFG

This chapter discusses how meaning has been handled in Lexical-Functional Grammar. As well as giving a historical overview, it also argues that the modern approach, using Glue Semantics, has a number of undesirable properties: meanings do not figure at all in the architecture of the grammar, merely standing in an unspecified correspondence with semantic structures; s-structures themselves have become an enfeebled and unimportant part of the projection architecture; and meaning constructors, when written in the ‘new glue’ format, give the impression of being quite distinct from other kinds of functional annotation, making the semantic component seem out of sync with the rest of the formalism. To remedy these issues, Findlay suggests a mechanism for representing meanings explicitly in s-structures, and for integrating linear logic into the description language of LFG more generally. This has immediate benefits for the analysis of idioms and for the theory of the semantics-information structure interface.

Music Stimulation as a Method of Optimizing Autobiographical Memory in Patients Diagnosed with Alzheimer’s Disease

Alzheimer’s is a neurodegenerative disease characterized by the progressive deterioration of cognitive functions, with memory being the most affected. Several studies have shown the benefits of music as a complementary treatment for dementia, improving patients’ quality of life. A scientific contribution is needed to show how autobiographic memory could be improved by using musical activities. Objective: The aim of this investigation is to analyze the impact of a musical stimulation protocol on the performance of autobiographical memory in elderly people suffering from Alzheimer’s. Participants and Method: This research was conducted with three patients diagnosed with Alzheimer’s disease: two females (66.7%), and one male (33.3%). One (33.3%) was in the early stages, and two were in the middle stages. This investigation used a quantitative, pre-experimental, longitudinal study with the application of two tests before and after the intervention. Findings: Changes in the performance of autobiographical memory (t=-5.79, p=0.002), and in the semantic component (t=-10.14, p=0.01) were found to be statistically significant, but no changes were evident for episodic memory (t=-0.19, p=0.86). Conclusion: This study provides preliminary evidence of the potential effectiveness of using a music protocol to improve the performance of autobiographical memory in patients with Alzheimer's Disease. Doi: 10.28991/esj-2021-01304 Full Text: PDF

STRUCTURAL AND SEMANTIC PECULIARITIES OF EMOTIONAL STATES VERBALISATION IN MODERN ENGLISH LITERATURE DISCOURSE (BASED ON THE NOVEL BY E. GILBERT ‘EAT, PRAY, LOVE’)

The article is devoted to the study of means expressing emotional states in modern English literature discourse, i.e., the novel by E. Gilbert ‘Eat, Pray, Love’. The paper outlines structural and semantic peculiarities of lexical means expressing emotional states based on 163 verbal constructions obtained by continuous sampling from the novel under analysis. Emotional and emotive vocabulary are two adjacent but not identical notions; the term ‘emotive’ belongs to the sphere of linguistics comprising multilevel lingual means of expressing emotions, i.e., a linguistic research focuses on emotive vocabulary where the verb plays a significant role as a meaningful centre of the sentence. Verbalisation of emotional states is performed synthetically (indication of the emotional state is incorporated into the verb semantics) and analytically (emotional state is expressed through an adjective or / and noun semantics which follow the link verb). Emotional states include means of positive, negative, and neutral states. Emotive vocabulary denoting positive emotions comprises 8 emotional states among which are love, gaiety, happiness, satisfaction, excitement, admiration, tranquillity, inspiration, with the units containing semantic component ‘love’ dominating this group and constructions denoting excitement, admiration, tranquillity, and inspiration demonstrating low productivity. Negative emotions include 7 emotional states: fear, grief, anxiety, anger, sadness, rage, anguish, where constructions denoting fear and grief demonstrate the highest productivity, and units denoting anxiety, anger, sadness, rage, and anguish show low productivity.

Axiological Component of Paroemias: Specificity and Role in Intercultural Communication

The article examines the axiological component of verbal units that comprise the paremiological funds in languages with different structure. The content of proverbs and sayings reflects cultural specificity along with the values adopted in a particular ethnic community. The analysis of 865 examples presented in the Russian, English, French and Latin languages has confirmed the hypothesis that the values of people are vividly reflected in paroemias. The author’s conception of “axiological component” has been clarified. Axiological component is understood as the value associated with the assessment included into the semantics of a paremiological unit. The semantic component of paroemias manifests itself in determining the value of an object or phenomenon in relation to other objects or phenomena. Also, it may reflect the work of categorization and classification mechanisms. The idea of axiological component correlation with the hierarchy of the value system (described by one of the founders of axiology H. Rickert) predetermines the novelty of this research. It is concluded that paroemias can be classified based on the content of the axiological component. The following groups have been singled out: behavioral instruction, basic life values, object assessment, subject assessment and assessment of the subject’s behavior. The didactics of paroemias can be expressed in explicit and implicit forms, but one way or another, it is present in every verbal unit. The value component is a serious obstacle in successful interaction; therefore, a projective analysis of the functioning of paroemias under the conditions of intercultural communication has been carried out.