scholarly journals Neural Linguistic Steganalysis via Multi-Head Self-Attention

2021 ◽  
Vol 2021 ◽  
pp. 1-5
Author(s):  
Sai-Mei Jiao ◽  
Hai-feng Wang ◽  
Kun Zhang ◽  
Ya-qi Hu
Keyword(s):  
Semantic Space ◽  
Multimedia Security ◽  
Semantic Features ◽  
Input Text

Linguistic steganalysis can indicate the existence of steganographic content in suspicious text carriers. Precise linguistic steganalysis on suspicious carrier is critical for multimedia security. In this paper, we introduced a neural linguistic steganalysis approach based on multi-head self-attention. In the proposed steganalysis approach, words in text are firstly mapped into semantic space with a hidden representation for better modeling the semantic features. Then, we utilize multi-head self-attention to model the interactions between words in carrier. Finally, a softmax layer is utilized to categorize the input text as cover or stego. Extensive experiments validate the effectiveness of our approach.

The Cluster Configuration of the Semantic Space of Polysituational Juridical Verbs with the Subject «Judge» in English

2020 ◽  
pp. 101-115
Author(s):  
Anton Silnitsky
Keyword(s):  
Cluster Analysis ◽  
Semantic Space ◽  
Semantic Feature ◽  
Negative Evaluation ◽  
Semantic Features ◽  
Structural Elements ◽  
Social Sphere ◽  
Quantitative Linguistics ◽  
Cluster Configuration ◽  
The Subject

The article is dedicated to the analysis of the semantic space of polysituational juridical verbs with the subject «judge» in English. The theoretical part of the research combines some aspects of «verb-centric» conception and «quantitative linguistics». A polisituational verb» implies a complex verbal situation consisting of several simple situations. The notion of a «juridical verb» correlates with a juridical social sphere. The article substantiates diagnostic semantic features which constitute the structural elements of the semantic space. The semantic features are organized into semantic plans. One semantic plan («chronostructural») being «categorical» integrates the features «basic situation» and «background situation». The «subcategorical» semantic plans are: «teleological » (the features: «definite» and «indefinite» situations), «temporal» (the features: «retrospective» and «prospective» situations), «motivational» («strong» and «weak» situations), «adject-legal» («base», «sanctions» and «processual» law), «adject-functional» («instrumental» and «purpose-oriented» law), «adjectsubstantive» («material» and «person» adject), «interactive» («convergent», «divergent » and «invergent»), «hierarchical» («dominant» and «subordinate» subject), «axiological» («positive» and «negative» evaluation). Each subcategorical semantic feature (exept for temporal plan features) correlates with one of the categorical («basic situation» or «background situation»). The actant «judge» is modeled by means of the following features within a basic situation: «processual» law, «instrumental» law and «dominant» subject. By means of cluster analysis the semantic features were grouped into two clusters («Divergent» and «Convergent-invergent») in correlation with an accusatorial or a justificatory-undifferentiated sentence. The differential (most relevant) semantic characteristics within the basic situation are the features of interactive and motivational plans, within the background situation they constitute the features of adject-legal, adject-functional and interactive plans.

scholarly journals Latent Semantics Encoding for Label Distribution Learning

2019 ◽  
Author(s):  
Suping Xu ◽  
Lin Shang ◽  
Furao Shen
Keyword(s):  
Empirical Studies ◽  
Semantic Space ◽  
Learning Ability ◽  
Semantic Features ◽  
Risk Minimization ◽  
Negative Effects ◽  
Real World Data ◽  
Empirical Risk ◽  
Label Distribution Learning ◽  
Label Distribution

Label distribution learning (LDL) is a newly arisen learning paradigm to deal with label ambiguity problems, which can explore the relative importance of different labels in the description of a particular instance. Although some existing LDL algorithms have achieved better effectiveness in real applications, most of them typically emphasize on improving the learning ability by manipulating the label space, while ignoring the fact that irrelevant and redundant features exist in most practical classification learning tasks, which increase not only storage requirements but also computational overheads. Furthermore, noises in data acquisition will bring negative effects on the generalization performance of LDL algorithms. In this paper, we propose a novel algorithm, i.e., Latent Semantics Encoding for Label Distribution Learning (LSE-LDL), which learns the label distribution and implements feature selection simultaneously under the guidance of latent semantics. Specifically, to alleviate noise disturbances, we seek and encode discriminative original physical/chemical features into advanced latent semantic features, and then construct a mapping from the encoded semantic space to the label space via empirical risk minimization. Empirical studies on 15 real-world data sets validate the effectiveness of the proposed algorithm.

scholarly journals Referentiality of event in postmodern fictional discourse: pragmatics and semantics

2021 ◽  
pp. 181-190
Author(s):  
Mariya I. Kudryavtseva
Keyword(s):  
Semantic Space ◽  
Formal Logic ◽  
Semantic Features ◽  
The Novel ◽  
Pragmatic Markers ◽  
The World ◽  
Discourse Pragmatics

The article is devoted to the study of postmodern fictional discourse referentialism in terms of pragmatics and semantics. Postmodern fictional discourse eliminates the oppositions of different narrative perspectives, which entails a non-distinction of the author’s narrative and the characters’ speech. The relevance of the article arises from the need to identify the communicative parameters of the creator of fictional discourse and its recipient from the standpoint of the cognitive and discursive linguistic paradigm, but it should be noted that both sides of fictional communication are free in the choice and interpretation of language elements outside the fixed conventional meanings. This feature of fictional discourse results in a set of conventions adopted by both sides of aesthetic communication. Metaphorically polysemantic images, which conflict with the definitions of formal logic and affirm the variability of the world and consciousness, are characteristic of postmodern fictional discourse. Pastiche, irony, collage, allusions, citations, direct derivations have significant pragmatic potential. The aim of the article is to identify and describe the pragmatic and semantic features of structuring the referentiality of an event in postmodern fictional discourse on the material of Sasha Sokolov’s novel “School for Fools”. The semantic space of the novel creates the dialogicity of the characters’ consciousness, pragmatically marked by dialogization of speech, which emphasizes the recipient’s attention to the imaginary event. The text fragments become independent discursive phenomena, thus contributing to the perception of the novel as a complex of macrocontexts. Markers of narrative polyphony, as well as intertextuality and precedent phenomena serve as pragmatic markers of the special referentiality of an event in S. Sokolov’s novel.

scholarly journals Progressive Domain-Independent Feature Decomposition Network for Zero-Shot Sketch-Based Image Retrieval

2020 ◽  
Author(s):  
Xinxun Xu ◽  
Muli Yang ◽  
Yanhua Yang ◽  
Hao Wang
Keyword(s):  
Image Retrieval ◽  
Semantic Information ◽  
Semantic Space ◽  
Semantic Knowledge ◽  
Visual Features ◽  
Semantic Features ◽  
Retrieval Task ◽  
Common Space ◽  
Low Dimensional ◽  
Domain Independent

Zero-Shot Sketch-Based Image Retrieval (ZS-SBIR) is a specific cross-modal retrieval task for searching natural images given free-hand sketches under the zero-shot scenario. Most existing methods solve this problem by simultaneously projecting visual features and semantic supervision into a low-dimensional common space for efficient retrieval. However, such low-dimensional projection destroys the completeness of semantic knowledge in original semantic space, so that it is unable to transfer useful knowledge well when learning semantic features from different modalities. Moreover, the domain information and semantic information are entangled in visual features, which is not conducive for cross-modal matching since it will hinder the reduction of domain gap between sketch and image. In this paper, we propose a Progressive Domain-independent Feature Decomposition (PDFD) network for ZS-SBIR. Specifically, with the supervision of original semantic knowledge, PDFD decomposes visual features into domain features and semantic ones, and then the semantic features are projected into common space as retrieval features for ZS-SBIR. The progressive projection strategy maintains strong semantic supervision. Besides, to guarantee the retrieval features to capture clean and complete semantic information, the cross-reconstruction loss is introduced to encourage that any combinations of retrieval features and domain features can reconstruct the visual features. Extensive experiments demonstrate the superiority of our PDFD over state-of-the-art competitors.

scholarly journals Computational reconstruction of mental representations using human behavior

2022 ◽  
Author(s):  
Laurent Caplette ◽  
Nicholas Turk-Browne
Keyword(s):  
Mental Representations ◽  
Semantic Space ◽  
Direct Access ◽  
Visual Features ◽  
Semantic Mapping ◽  
Semantic Features ◽  
Visual Concept ◽  
Visual Concepts ◽  
Computational Reconstruction ◽  
High Level

Revealing the contents of mental representations is a longstanding goal of cognitive science. However, there is currently no general framework for providing direct access to representations of high-level visual concepts. We asked participants to indicate what they perceived in images synthesized from random visual features in a deep neural network. We then inferred a mapping between the semantic features of their responses and the visual features of the images. This allowed us to reconstruct the mental representation of virtually any common visual concept, both those reported and others extrapolated from the same semantic space. We successfully validated 270 of these reconstructions as containing the target concept in a separate group of participants. The visual-semantic mapping uncovered with our method further generalized to new stimuli, participants, and tasks. Finally, it allowed us to reveal how the representations of individual observers differ from each other and from those of neural networks.

Plagiarism Detection Using Machine Learning-Based Paraphrase Recognizer

2016 ◽  
Vol 25 (3) ◽  
pp. 351-359 ◽  
Author(s):  
A. Chitra ◽  
Anupriya Rajkumar
Keyword(s):  
Recognition System ◽  
Support Vector ◽  
Free Text ◽  
Semantic Features ◽  
Plagiarism Detection ◽  
Detection Systems ◽  
Passage Level ◽  
Input Text ◽  
Sentence Level ◽  
Wide Availability

AbstractPlagiarism in free text has become a common occurrence due to the wide availability of voluminous information resources. Automatic plagiarism detection systems aim to identify plagiarized content present in large repositories. This task is rendered difficult by the use of sophisticated plagiarism techniques such as paraphrasing and summarization, which mask the occurrence of plagiarism. In this work, a monolingual plagiarism detection technique has been developed to tackle cases of paraphrased plagiarism. A support vector machine based paraphrase recognition system, which works by extracting lexical, syntactic, and semantic features from input text has been used. Both sentence-level and passage-level approaches have been investigated. The performance of the system has been evaluated on various corpora, and the passage level approach has registered promising results.

scholarly journals Conceptual and Perceptual Dimensions of Word Meaning Are Recovered Rapidly and in Parallel during Reading

2019 ◽  
Vol 31 (1) ◽  
pp. 95-108
Author(s):  
Valentina Borghesani ◽  
Marco Buiatti ◽  
Evelyn Eger ◽  
Manuela Piazza
Keyword(s):  
Word Reading ◽  
Temporal Dynamics ◽  
Brain Activity ◽  
Word Meaning ◽  
Low Frequency ◽  
Semantic Category ◽  
Semantic Space ◽  
Semantic Features ◽  
Healthy Human ◽  
Low Frequency Oscillations

A single word (the noun “ elephant”) encapsulates a complex multidimensional meaning, including both perceptual (“ big”, “ gray”, “ trumpeting”) and conceptual (“ mammal”, “ can be found in India”) features. Opposing theories make different predictions as to whether different features (also conceivable as dimensions of the semantic space) are stored in similar neural regions and recovered with similar temporal dynamics during word reading. In this magnetoencephalography study, we tracked the brain activity of healthy human participants while reading single words varying orthogonally across three semantic dimensions: two perceptual ones (i.e., the average implied real-world size and the average strength of association with a prototypical sound) and a conceptual one (i.e., the semantic category). The results indicate that perceptual and conceptual representations are supported by partially segregated neural networks: Whereas visual and auditory dimensions are encoded in the phase coherence of low-frequency oscillations of occipital and superior temporal regions, respectively, semantic features are encoded in the power of low-frequency oscillations of anterior temporal and inferior parietal areas. However, despite the differences, these representations appear to emerge at the same latency: around 200 msec after stimulus onset. Taken together, these findings suggest that perceptual and conceptual dimensions of the semantic space are recovered automatically, rapidly, and in parallel during word reading.

scholarly journals SEMANTIC FEATURES OF THE PHRASEOLOGICAL UNITS WITH THE COMPONENT LIGHT

2019 ◽  
Vol 7 (4) ◽  
pp. 33-38
Author(s):  
Irina A. Kuprieva ◽  
Stanislava B. Smirnova ◽  
Ludmila V. Belova ◽  
Vladimir S. Pugach
Keyword(s):  
Semantic Analysis ◽  
Comparative Method ◽  
Semantic Space ◽  
Semantic Features ◽  
Continuous Sampling ◽  
Postmodern Fiction ◽  
Study Results

Purpose: To conduct lexical and semantic analysis on the concept light in the artistic discourse of postmodern fiction. Methodology: In this research, comparative method, search for synonyms, continuous sampling and seminal analysis are used. Main Findings: As a result, it becomes obvious that the boundaries of the artistic concepts are extremely blurred and much wider than the boundaries of the corresponding non-artistic concepts. In conclusion, the concept light is frequently used in English artistic discourses, i.e. significant in the culture, and a special status in the semantic space of the concept light has the meaning of knowledge and information. Applications: The study results can be used by students and universities. Novelty/Originality: In this research, a model of the semantic features of the phraseological units with the component light is presented in a comprehensive and complete manner.

Sign in / Sign up

Export Citation Format

Share Document