Semantic Concepts Research Articles

An incremental clustering algorithm based on semantic concepts

An incremental clustering algorithm based on semantic concepts

Envisioning the Semantics of Kinyarwanda causative and applicative morphemes in the Cognitive Grammar Theory

This paper uses the Cognitive Grammar Theory to examine the semantics of Kinyarwanda causative and applicative. The study was motivated by the fact that Kinyarwanda verb extensions, particularly the causative and applicative, have multiple sociocultural interpretations of meanings, which is a typical characteristic of Cognitive semantics. The study was couched in the interpretivism paradigm, which looks for culturally derived interpretations historically situated in the surroundings of humans. A qualitative approach was used to analyse the meanings of Kinyarwanda words used by Nyarwanda native speakers. Data was collected from focus group discussions involving six Kinyarwanda speakers. Four were monolinguals; the rest were multilingual, speaking Kiswahili, English and Kinyarwanda. The data were presented using Leipzig glossing rules, and illustrations were used to show more than one meaning in the derived verbal lexemes. Thus, data analysis was used using cognitive grammar theoretical apparatuses, which account for multiple meanings of the derived lexemes. The study found that the causative and applicative verb extensions attached to the verb root have multiple senses that line with Cognitive Grammar Theory, whose major assumption is that words have multiple conceptualisations. For instance, the derived verb vugisha had four meanings: speak to, make or cause to talk, disgusting and switching on something (such as in radio receiver or electricity). As the presence of different meanings in the same word causes ambiguity, further research can be carried out on Kinyarwanda ambiguity under the contemporary theory of Metaphor.

Cognitive Semantics and The Related Theory

يهدف هذا المقال إلى تسليط الضوء على دراسة الدلالة المعرفية ومدى تأثيرها في دراسة المعنى، إذ ترتبط ارتباطا وثيقا بعدد من النظريات من بينها العمليات التفسيرية. تفيد هذه المقالة في معرفة الأسس الأساسية لعلم الدلالة وكيفية تعامله مع المعنى الذي يتقاطع بشكل مباشر بين المتكلم والمجتمع. ولذلك سيتم التركيز في هذا المقال بشكل أساسي على دراسة الدلالات المعرفية ودورها وأساسياتها، بالإضافة إلى معرفة إحدى النظريات المهمة في دراسة المعنى وفروعه الرئيسية. وبعد ذلك نستنتج من خلال هذا المقال أن علم الدلالة المعرفي هو جانب أساسي يهتم بدراسة المعنى الموجود في ذهن المتكلم. إنهم مرتبطون ارتباطًا وثيقًا بالمجتمع، ويرتبطون بالإدراك. الكلمات المفتاحية: علم المعنى الإدراكي , الوسائل العملية ,الانتباه , المقارنة

Learning Relationship-Enhanced Semantic Graph for Fine-Grained Image-Text Matching.

Image-text matching of natural scenes has been a popular research topic in both computer vision and natural language processing communities. Recently, fine-grained image-text matching has shown its significant advance in inferring the high-level semantic correspondence by aggregating pairwise region-word similarity, but it remains challenging mainly due to insufficient representation of high-order semantic concepts and their explicit connections in one modality as its matched in another modality. To tackle this issue, we propose a relationship-enhanced semantic graph (ReSG) model, which can improve the image-text representations by learning their locally discriminative semantic concepts and then organizing their relationships in a contextual order. To be specific, two tailored graph encoders, visual relationship-enhanced graph (VReG) and textual relationship-enhanced graph (TReG), are respectively exploited to encode the high-level semantic concepts of corresponding instances and their semantic relationships. Meanwhile, the representations of each graph node are optimized by aggregating semantically contextual information to enhance the node-level semantic correspondence. Further, the hard-negative triplet ranking loss, center hinge loss, and positive-negative margin loss are jointly leveraged to learn the fine-grained correspondence between the ReSG representations of image and text, whereby the discriminative cross-modal embeddings can be explicitly obtained to benefit various image-text matching tasks in a more interpretable way. Extensive experiments verify the advantages of the proposed fine-grained graph matching approach, by achieving the state-of-the-art image-text matching results on public benchmark datasets.

Measuring semantic memory using associative and dissociative retrieval tasks.

Recent theoretical advances highlighted the need for novel means of assessing semantic cognition. Here, we introduce the associative-dissociative retrieval task (ADT), positing a novel way to test inhibitory control over semantic memory retrieval by contrasting the efficacy of associative (automatic) and dissociative (controlled) retrieval on a standard set of verbal stimuli. All ADT measures achieved excellent reliability, homogeneity, and short-term temporal stability. Moreover, in-depth stimulus level analyses showed that the associative retrieval is easier for words evoking few but strong associates, yet such propensity hampers the inhibition. Finally, we provided critical support for the construct validity of the ADT measures, demonstrating reliable correlations with domain-specific measures of semantic memory functioning (semantic fluency and associative combination) but negligible correlations with domain-general capacities (processing speed and working memory). Together, we show that ADT provides simple yet potent and psychometrically sound measures of semantic memory retrieval and offers noteworthy advantages over the currently available assessment methods.

The Notion of Image Schemas in Kurdish Proverbs

This study examines the notion of image schemas in selected Kurdish proverbs. As used in Cognitive Semantics, image schemas refer to conceptual structures that represent recurring patterns in our experience of the physical and psychological world. They are abstract concepts consisting of patterns emerging from repeated instances of embodied experience. Examples of image schemas include CONTAINER, PATH, FORCE, SCALE and CYCLE schemas. Applied to Kurdish proverbs, it is argued that image schemas serve as the basis for organizing knowledge and reasoning about the world. They are derived from concrete physical experiences that are projected onto abstract concepts. The aim of the study then is to show how image schemas provide the basis for richly detailed lexical concepts. One interesting finding is that image schemas arise directly from sensory and perceptual experience. They are functions of Kurdish speakers’ everyday interaction with and observation of the world around them.

In search of methodological standards for corpus-based cognitive semantics: The case of Behavioral Profiles

ABSTRACT As a result of the quantitative turn, corpus data have become a staple of cognitive semantic research, and the strong emphasis on issues of methodology has led to a proliferation of analytical approaches. Successful application of any one approach requires a thorough understanding of theoretically-grounded methodological requirements and their implications. The present paper focuses on the framework of Behavioral Profiles and offers a methodologically-oriented review of nine studies carried out by the original authoring team and collaborating researchers. Through close scrutiny of the research process, it specifically aims to outline and examine the methodological standards required for data collection, annotation and statistical analysis. By discussing a range of available choices and highlighting examples of best practice, this paper will hopefully facilitate informed decision-making in future research into the organization of near-synonymous and polysemous categories.

Semantic classifiers in Guodian bamboo manuscripts: Reconstructing categories in the ancient Chinese mind

Ancient Chinese scripts are one of the most complex writing systems in the world. Many graphs contain components referred to as unpronounced semantic classifiers. These components bear a semantic relationship to the meaning of the word written by the graph. It is proposed that the set of semantic components, as well as the specific semantic concepts that they encode, reflect a cognitive map of knowledge organization of the community of script users at a particular place and time. By analyzing the semantic classifiers within a constrained manuscript corpus, it is possible to build a network that models aspects of the ancient Chinese knowledge organization as reflected in this corpus. With the help of the digital tool iClassifier, a corpus-based study is conducted based on the sample texts, Guodian bamboo manuscripts. Additionally, a list of semantic classifiers is presented, two prominent categories 心 [heart/senses & emotion] and 辵 [road + foot/movement] are analyzed, and a visualization network of the sample corpus is drawn.

Transcranial direct current stimulation over the left posterior temporal lobe modulates semantic control: Evidence from episodic memory distortions

Evidence accumulates to show that semantic cognition requires, in addition to semantic representations, control processes that regulate the accessibility and use of semantic knowledge in a task- and time-appropriate fashion. Semantic control has been recently proposed to rely on a distributed network that includes the posterior temporal cortex. Along these lines, recent meta-analyses of neuroimaging data and studies with patients suffering from semantic aphasia have suggested that the left posterior middle temporal gyrus (pMTG) is critically involved whenever situational context must constrain semantic retrieval. In the present experiment, we used transcranial direct current stimulation over the left posterior temporal lobe in an attempt to interfere with semantic control while participants performed a DRM task, a procedure for inducing conceptually-based false recognition that is contingent on both activation and control processes. Paralleling findings with patients suffering from brain damage restricted to the temporoparietal cortex, anodal stimulation (relative to sham stimulation) resulted in increased false recognition but intact true recognition. These findings fit well with the idea that the left pMTG is a key component of a semantic control network, the alteration of which results in memory performance that is affected by the intrusion of contextually-inappropriate semantic information.

Nominal Recursors as Epi-Recursors

We study nominal recursors from the literature on syntax with bindings and compare them with respect to expressiveness. The term "nominal" refers to the fact that these recursors operate on a syntax representation where the names of bound variables appear explicitly, as in nominal logic. We argue that nominal recursors can be viewed as epi-recursors, a concept that captures abstractly the distinction between the constructors on which one actually recurses, and other operators and properties that further underpin recursion. We develop an abstract framework for comparing epi-recursors and instantiate it to the existing nominal recursors, and also to several recursors obtained from them by cross-pollination. The resulted expressiveness hierarchies depend on how strictly we perform this comparison, and bring insight into the relative merits of different axiomatizations of syntax. We also apply our methodology to produce an expressiveness hierarchy of nominal corecursors, which are principles for defining functions targeting infinitary non-well-founded terms (which underlie lambda-calculus semantics concepts such as Böhm trees). Our results are validated with the Isabelle/HOL theorem prover.

Development of Few-Shot Learning Capabilities in Artificial Neural Networks When Learning Through Self-Supervised Interaction.

Most artificial neural networks used for object recognition are trained in a fully supervised setup. This is not only resource consuming as it requires large data sets of labeled examples but also quite different from how humans learn. We use a setup in which an artificial agent first learns in a simulated world through self-supervised, curiosity-driven exploration. Following this initial learning phase, the learned representations can be used to quickly associate semantic concepts such as different types of doors using one or more labeled examples. To do this, we use a method we call fast concept mapping which uses correlated firing patterns of neurons to define and detect semantic concepts. This association works instantaneously with very few labeled examples, similar to what we observe in humans in a phenomenon called fast mapping. Strikingly, we can already identify objects with as little as one labeled example which highlights the quality of the encoding learned self-supervised through interaction with the world. It therefore presents a feasible strategy for learning concepts without much supervision and shows that through pure interaction meaningful representations of an environment can be learned that work better for few-short learning than non-interactive methods.

Polarity Loss: Improving Visual-Semantic Alignment for Zero-Shot Detection.

Conventional object detection models require large amounts of training data. In comparison, humans can recognize previously unseen objects by merely knowing their semantic description. To mimic similar behavior, zero-shot object detection (ZSD) aims to recognize and localize "unseen" object instances by using only their semantic information. The model is first trained to learn the relationships between visual and semantic domains for seen objects, later transferring the acquired knowledge to totally unseen objects. This setting gives rise to the need for correct alignment between visual and semantic concepts so that the unseen objects can be identified using only their semantic attributes. In this article, we propose a novel loss function called "polarity loss" that promotes correct visual-semantic alignment for an improved ZSD. On the one hand, it refines the noisy semantic embeddings via metric learning on a "semantic vocabulary" of related concepts to establish a better synergy between visual and semantic domains. On the other hand, it explicitly maximizes the gap between positive and negative predictions to achieve better discrimination between seen, unseen, and background objects. Our approach is inspired by embodiment theories in cognitive science that claim human semantic understanding to be grounded in past experiences (seen objects), related linguistic concepts (word vocabulary), and visual perception (seen/unseen object images). We conduct extensive evaluations on the Microsoft Common Objects in Context (MS-COCO) and Pascal Visual Object Classes (VOC) datasets, showing significant improvements over state of the art. Our code and evaluation protocols available at: https://github.com/salman-h-khan/PL-ZSD_Release.

On Exploring Multiplicity of Primitives and Attributes for Texture Recognition in the Wild.

Texture recognition is a challenging visual task since its multiple primitives or attributes can be perceived from the texture image under different spatial contexts. Existing approaches predominantly built upon CNN incorporate rich local descriptors with orderless aggregation to capture invariance to the spatial layout. However, these methods ignore the inherent structure relation organized by primitives and the semantic concept described by attributes, which are critical cues for texture representation. In this paper, we propose a novel Multiple Primitives and Attributes Perception network (MPAP) that extracts features by modeling the relation of bottom-up structure and top-down attribute in a multi-branch unified framework. A bottom-up process is first proposed to capture the inherent relation of various primitive structures by leveraging structure dependency and spatial order information. Then, a top-down process is introduced to model the latent relation of multiple attributes by transferring attribute-related features between adjacent branches. Moreover, an augmentation module is devised to bridge the gap between high-level attributes and low-level structure features. MPAP can learn representation through jointing bottom-up and top-down processes in a mutually reinforced manner. Experimental results on six challenging texture datasets demonstrate the superiority of MPAP over state-of-the-art methods in terms of accuracy, robustness, and efficiency.

Semantičko-pojmovna analiza frazema sa komponentom život u srpskom, francuskom i italijanskom jeziku

Idioms provide important material for cross-linguistic and cross-cultural studies, as they reflect both a universal way of thinking inherent in different linguistic cultures and the peculiarities of a given society and culture. The subject of this research are Serbian, French and Italian idioms containing the lexemes (Serb.) život / (Fr.) vie / (It.) vita. Idioms are excerpted from several general and phraseological dictionaries of the Serbian, French and Italian languages, daily and weekly newspapers, and electronic media. The examples are grouped into conceptual fields according to the background images that motivate their meanings. The aim is to identify the similarities and differences in the con- cepts to which the selected idioms refer in the observed languages, as well as to determine the extent to which the concept of life motivates positive phraseological meanings, and the extent to which it contributes to the negative semantics of idioms. Appropriate descrip- tors that represent similarities between idioms are attached to the excerpted examples us- ing the method of semantic decomposition. Corpus analysis confirms the anthropocentric character of phraseology, which is reflected in the fact that the majority of idioms are related to human behavior, characteristics, states and good and bad behaviours toward others. Furthermore, the findings of this research show certain specificities regarding the phraseological expression of speakers of the observed languages, while a large number of equivalent examples, i.e. idioms of the same semantic concept, indicate similarities in the perception of life in the Serbian, French, and Italian linguistic cultures, as one of the basic, fundamental concepts.

Transformer networks of human conceptual knowledge.

We present a computational model capable of simulating aspects of human knowledge for thousands of real-world concepts. Our approach involves a pretrained transformer network that is further fine-tuned on large data sets of participant-generated feature norms. We show that such a model can successfully extrapolate from its training data, and predict human knowledge for new concepts and features. We apply our model to stimuli from 25 previous experiments in semantic cognition research and show that it reproduces many findings on semantic verification, concept typicality, feature distribution, and semantic similarity. We also compare our model against several variants, and by doing so, establish the model properties that are necessary for good prediction. The success of our approach shows how a combination of language data and (laboratory-based) psychological data can be used to build models with rich world knowledge. Such models can be used in the service of new psychological applications, such as the modeling of naturalistic semantic verification and knowledge retrieval, as well as the modeling of real-world categorization, decision-making, and reasoning. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

NewAbstractConcepts: A Database of 42 Normed Abstract Concepts and Exemplars.

Recently, researchers have expressed challenges in conducting word-learning experiments in adult populations due to limited availability of normed stimulus materials. This constraint often prompts the use of low-frequency or low-prevalence words, introducing the potential influence of prior knowledge or direct translation to familiar words. In response, we developed novel abstract concepts devoid of word referents, providing better control over prior knowledge. These new concepts describe situations encountered in various settings for which there is no existing word in English. The resulting database comprises 42 normed New Abstract Concepts, offering unique materials structured through scenarios, each containing similar and dissimilar exemplars. These materials underwent meticulous norming for relatability and similarity levels across a series of studies. The success of our approach was demonstrated in a word-learning experiment examining the effects of similarity and diversity. The database serves as a valuable resource for selecting stimuli in experiments exploring the learning of abstract semantic concepts, particularly investigating the role of similarity versus diversity in concept learning. The database is available on OSF (https://osf.io/svm2p/).

Predicting Semantic Concepts in Heterogeneous Radiographic Images

Predicting Semantic Concepts in Heterogeneous Radiographic Images

Model-Guided Generative Adversarial Networks for Unsupervised Fine-Grained Image Generation

Unsupervised fine-grained image generation is a challenging issue in computer vision. Although many recent significant advances have improved performance, the ability to synthesize photo-realistic images in an unsupervised manner remains extremely difficult. The existing methods compose an image via complex three-stage generative adversarial networks and impose constraints between the latent codes. This pipeline focuses on the disentanglement and ignores the quality of generated images. In this article, we propose a novel two-stage approach for unsupervised fine-grained image generation, termed Model-Guided Generative Adversarial Networks (MG-GAN). We introduce an attention module for exploring the correlation between fine-grained latent codes and image features in the foreground generation stage. The attention module enables the network to automatically focus on the color details and semantic concepts of objects related to different fine-grained classes. Furthermore, we incorporate knowledge distillation strategy and design a simple but effective inverse background image generator as a teacher to guide the background image generation. With the help of knowledge learned in the pre-trained inverse background image generator, a comfortable canvas is synthesized and combined with foreground object more reasonably. Extensive experiments on three popularly fine-grained datasets demonstrate that our approach achieves state-of-the-art performance and is even competitive with semi-supervised method.

Dynamic Color Design Model Construction Considering Ambient Light Changes

Abstract The study takes dynamic color design as the main research object, firstly, based on the atmospheric scattering model and the improved dark primary color a priori algorithm, designing the image defogging algorithm based on the ambient light variation, which is used to carry out the optimization processing of images. Aiming at the lack of the user’s visual cognitive semantics of color matching in the process of color design, it is proposed to construct a dynamic color design model based on the theoretical foundation of the image translation model, combined with the visual aesthetics data flow. The model obtains the user’s visual aesthetic cognition of the optimized image samples through eye tracking technology and, at the same time, applies the image translation model to the color design process to generate a color design scheme that is more in line with the human visual characteristics by learning the color matching law of the samples and their visual cognitive semantics. The experimental analysis shows that the image-defogging algorithm based on ambient light variation has a better image-processing effect and operation efficiency, exhibiting a higher average gradient (11.83), information entropy (7.86) and visual fidelity (2.99). Meanwhile, the color design scheme generated by the method of this paper is highly recognized by the subjects, with evaluation scores of more than 4 points in each dimension. The dynamic color design model constructed in this paper, by considering ambient light changes, can generate color schemes that are both rational and comfortable.

Сognitive Dimension of the Word-forming Nest with Dialectal Vertices Yeden / Yeden (“єде́н / є́ден”): Logical Propositions

In this article, a current task of modern derivatolofical studies is addressed – the cognitive structuring of derived words in denumerative word formation, grouped according to word-forming semantics, considering motivational relationships through the prism of divergent-convergent properties of the content of the generative word – the vertex of the word-formation nest. The analysis was carried out on the basis of a word-forming nest with variational vertices belonging to the dialect layer of the lexicon – yeden /yeden (“еден / еден”). The specified motivational words are characterized by a polysemantic continuum, so they are characterized by differences in creative meanings both directly and indirectly deep into the word-forming nest. Such complex units of word-formation have the status of being divergent. Motivating relations in the dyad «the vertex of the word-forming nest – the denumerative» can record mental knowledge in the aspect of the linguo-cognitive approach to the analysis of linguistic phenomena: the genetic codes of the word-formation, preserved in the derivative, as well as the semantic increments present in the meaning structure of the derivatives are revealed. Analyzing the cognitive semantics within word-forming values, the elements of frame are established, the components of which are derivative units of modification and transposition types of derivation, namely: subframe, cognitive slot, verbalized slot, propositional structure, and two-component logical proposition. The prospects of the research consist in the systematіc cognitive interpretation of derivatives based on various word-formation units, which lead the formatives that belong not only to the core, but also to the periphery of the motivational base, which will make it possible to comprehensively model the semantic positions of derived words, or more broadly, the mental manifestation of the word-formation level Ukrainian language.