Semantic Knowledge Representation Research Articles

Enhanced generalization and specialization of brain representations of semantic knowledge in healthy aging

Aging is often associated with a decrease in cognitive capacities. However, semantic memory appears relatively well preserved in healthy aging. Both behavioral and neuroimaging studies support the view that changes in brain networks contribute to this preservation of semantic cognition. However, little is known about the role of healthy aging in the brain representation of semantic categories. Here we used pattern classification analyses and computational models to examine the neural representations of living and non-living word concepts. The results demonstrate that brain representations of animacy in healthy aging exhibit increased similarity across categories, even across different task contexts. This pattern of results aligns with the neural dedifferentiation hypothesis that proposes that aging is associated with decreased specificity in brain activity patterns and less efficient neural resource allocation. However, the loss in neural specificity for different categories was accompanied by increased dissimilarity of item-based conceptual representations within each category. Taken together, the age-related patterns of increased generalization and specialization in the brain representations of semantic knowledge may reflect a compensatory mechanism that enables a more efficient coding scheme characterized by both compression and sparsity, thereby helping to optimize the limited neural resources and maintain semantic processing in the healthy aging brain.

Research on the construction and mapping model of knowledge organization system driven by standards

With the rapid development of artificial intelligence and enterprise digital transformation, the standardization organization, storage and management of semantic knowledge in computers have become the current research focus. As the core theory of knowledge system construction, knowledge organization (KO) provides theoretical support for the study of semantic knowledge organization and representation, among which knowledge organization system (KOS) is the important tool of semantic organization. At present, many scholars have carried out research from different perspectives of KOS based on theory, which provides the direction for the sustainable development of KOS. However, most of these studies focus on some aspects of KOS, which are in a "scattered" state, lacking systematic analysis of the basic principles of KOS construction and semantic organization based on theories and international standards. Therefore, this paper firstly constructs KOS theoretical models in the conceptual world and computer world respectively through a comprehensive study of multi-disciplinary basic theories such as semantics, logic, system theory, and international standards such as ISO 1087:2019, ISO 25964:2013, and ISO 11179:2023, and traces the iterative construction, organization and mapping process from "concept" in the conceptual world to "metadata" knowledge and semantics in the computer world. The semantic organization based on metadata is realized in computer. Secondly, on this basis, in order to realize ontology representation of domain knowledge, the ontology construction method based on MDR metadata is proposed. Finally, taking the semantic organization and ontology construction of Epicentre model in petroleum field as an example, the feasibility of the ideas and methods proposed in this paper is verified. The model and method proposed in this paper is independent of the specific type of KOS, so it is innovative and universal. The methodology is also applicable to other fields of conceptual system modeling, metadata standard construction, and data model modeling.

Involving logical clinical knowledge into deep neural networks to improve bladder tumor segmentation

Segmentation of bladder tumors from medical radiographic images is of great significance for early detection, diagnosis and prognosis evaluation of bladder cancer. Deep Convolution Neural Networks (DCNNs) have been successfully used for bladder tumor segmentation, but the segmentation based on DCNN is data-hungry for model training and ignores clinical knowledge. From the clinical view, bladder tumors originate from the mucosal surface of bladder and must rely on the bladder wall to survive and grow. This clinical knowledge of tumor location is helpful to improve the bladder tumor segmentation. To achieve this, we propose a novel bladder tumor segmentation method, which incorporates the clinical logic rules of bladder tumor and bladder wall into DCNNs to harness the tumor segmentation. Clinical logical rules provide a semantic and human-readable knowledge representation and are easy for knowledge acquisition from clinicians. In addition, incorporating logical rules of clinical knowledge helps to reduce the data dependency of the segmentation network, and enables precise segmentation results even with limited number of annotated images. Experiments on bladder MR images collected from the collaborating hospital validate the effectiveness of the proposed bladder tumor segmentation method.

PCAO2: an ontology for integration of prostate cancer associated genotypic, phenotypic and lifestyle data.

Disease ontologies facilitate the semantic organization and representation of domain-specific knowledge. In the case of prostate cancer (PCa), large volumes of research results and clinical data have been accumulated and needed to be standardized for sharing and translational researches. A formal representation of PCa-associated knowledge will be essential to the diverse data standardization, data sharing and the future knowledge graph extraction, deep phenotyping and explainable artificial intelligence developing. In this study, we constructed an updated PCa ontology (PCAO2) based on the ontology development life cycle. An online information retrieval system was designed to ensure the usability of the ontology. The PCAO2 with a subclass-based taxonomic hierarchy covers the major biomedical concepts for PCa-associated genotypic, phenotypic and lifestyle data. The current version of the PCAO2 contains 633 concepts organized under three biomedical viewpoints, namely, epidemiology, diagnosis and treatment. These concepts are enriched by the addition of definition, synonym, relationship and reference. For the precision diagnosis and treatment, the PCa-associated genes and lifestyles are integrated in the viewpoint of epidemiological aspects of PCa. PCAO2 provides a standardized and systematized semantic framework for studying large amounts of heterogeneous PCa data and knowledge, which can be further, edited and enriched by the scientific community. The PCAO2 is freely available at https://bioportal.bioontology.org/ontologies/PCAO, http://pcaontology.net/ and http://pcaontology.net/mobile/.

A Review of Immersive Technologies, Knowledge Representation, and AI for Human-Centered Digital Experiences

The evolution of digital technologies has resulted in the emergence of diverse interaction technologies. In this paper, we conducted a review of seven domains under a human-centric approach user interface design, human-centered web-based information systems, semantic knowledge representation, X-reality applications, human motion and 3D digitization, serious games, and AI. In this review, we studied these domains concerning their impact on the way we interact with digital interfaces, process information, and engage in immersive experiences. As such, we highlighted the shifts in design paradigms, user-centered principles, and the rise of web-based information systems. The results of such shifts are materialized in modern immersive technologies, semantic knowledge representation, serious games, and the facilitation of artificial intelligence for interactions. Through this exploration, we aimed to assist our understanding of the challenges that lie ahead. The seamless integration of technologies, ethical considerations, accessibility, education for technological literacy, interoperability, user trust, environmental sustainability, and regulatory frameworks are becoming significant. These challenges present opportunities for the future to enrich human experiences while addressing societal needs. This paper lays the groundwork for thoughtful and innovative approaches to the challenges that will define the future of human–computer interaction and information technologies.

Spatial versus graphical representation of distributional semantic knowledge.

Spatial distributional semantic models represent word meanings in a vector space. While able to model many basic semantic tasks, they are limited in many ways, such as their inability to represent multiple kinds of relations in a single semantic space and to directly leverage indirect relations between two lexical representations. To address these limitations, we propose a distributional graphical model that encodes lexical distributional data in a graphical structure and uses spreading activation for determining the plausibility of word sequences. We compare our model to existing spatial and graphical models by systematically varying parameters that contributing to dimensions of theoretical interest in semantic modeling. In order to be certain about what the models should be able to learn, we trained each model on an artificial corpus describing events in an artificial world simulation containing experimentally controlled verb-noun selectional preferences. The task used for model evaluation requires recovering observed selectional preferences and inferring semantically plausible but never observed verb-noun pairs. We show that the distributional graphical model performed better than all other models. Further, we argue that the relative success of this model comes from its improved ability to access the different orders of spatial representations with the spreading activation on the graph, enabling the model to infer the plausibility of noun-verb pairs unobserved in the training data. The model integrates classical ideas of representing semantic knowledge in a graph with spreading activation and more recent trends focused on the extraction of lexical distributional data from large natural language corpora. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Synthesizing Knowledge-Enhanced Features for Real-World Zero-Shot Food Detection.

Food computing brings various perspectives to computer vision like vision-based food analysis for nutrition and health. As a fundamental task in food computing, food detection needs Zero-Shot Detection (ZSD) on novel unseen food objects to support real-world scenarios, such as intelligent kitchens and smart restaurants. Therefore, we first benchmark the task of Zero-Shot Food Detection (ZSFD) by introducing FOWA dataset with rich attribute annotations. Unlike ZSD, fine-grained problems in ZSFD like inter-class similarity make synthesized features inseparable. The complexity of food semantic attributes further makes it more difficult for current ZSD methods to distinguish various food categories. To address these problems, we propose a novel framework ZSFDet to tackle fine-grained problems by exploiting the interaction between complex attributes. Specifically, we model the correlation between food categories and attributes in ZSFDet by multi-source graphs to provide prior knowledge for distinguishing fine-grained features. Within ZSFDet, Knowledge-Enhanced Feature Synthesizer (KEFS) learns knowledge representation from multiple sources (e.g., ingredients correlation from knowledge graph) via the multi-source graph fusion. Conditioned on the fusion of semantic knowledge representation, the region feature diffusion model in KEFS can generate fine-grained features for training the effective zero-shot detector. Extensive evaluations demonstrate the superior performance of our method ZSFDet on FOWA and the widely-used food dataset UECFOOD-256, with significant improvements by 1.8% and 3.7% ZSD mAP compared with the strong baseline RRFS. Further experiments on PASCAL VOC and MS COCO prove that enhancement of the semantic knowledge can also improve the performance on general ZSD. Code and dataset are available at https://github.com/LanceZPF/KEFS.

Supplemental Material for Spatial Versus Graphical Representation of Distributional Semantic Knowledge

Supplemental Material for Spatial Versus Graphical Representation of Distributional Semantic Knowledge

Semantic representation of neural circuit knowledge in Caenorhabditis elegans

In modern biology, new knowledge is generated quickly, making it challenging for researchers to efficiently acquire and synthesise new information from the large volume of primary publications. To address this problem, computational approaches that generate machine-readable representations of scientific findings in the form of knowledge graphs have been developed. These representations can integrate different types of experimental data from multiple papers and biological knowledge bases in a unifying data model, providing a complementary method to manual review for interacting with published knowledge. The Gene Ontology Consortium (GOC) has created a semantic modelling framework that extends individual functional gene annotations to structured descriptions of causal networks representing biological processes (Gene Ontology–Causal Activity Modelling, or GO–CAM). In this study, we explored whether the GO–CAM framework could represent knowledge of the causal relationships between environmental inputs, neural circuits and behavior in the model nematode C. elegans [C. elegans Neural–Circuit Causal Activity Modelling (CeN–CAM)]. We found that, given extensions to several relevant ontologies, a wide variety of author statements from the literature about the neural circuit basis of egg-laying and carbon dioxide (CO2) avoidance behaviors could be faithfully represented with CeN–CAM. Through this process, we were able to generate generic data models for several categories of experimental results. We also discuss how semantic modelling may be used to functionally annotate the C. elegans connectome. Thus, Gene Ontology-based semantic modelling has the potential to support various machine-readable representations of neurobiological knowledge.Graphical

A taxonomy for decision making in IoT systems

Semantic knowledge representations in the IoT can enable the vision of autonomic computing by providing a specification that enables interoperability and reasoning. Nevertheless, semantic representations in IoT have focused on describing the elements that compose it and their interactions, without addressing the challenges of the logical evolution of a system (updating and design of new algorithms). This work focuses on this gap, proposing the Taxonomy for Decision Making in IoT Systems (TDMIoT), a high-level characterization of decision-making processes in IoT developed following a conceptual-empirical methodological approach. TDMIoT considers a decision-making process a problem-solution association aiming to deliver a semantic representation that can be used as a design framework to support changes or even the design of new decision processes. A systematic review of the literature on decision-making processes in IoT application domains was conducted to evaluate the taxonomy as a classification scheme. A summary of the state-of-the-art decision-making process design approaches was generated from the classification of the selected studies through the systematic review. The classification showed design bias regarding the decision processes. For instance, most studies have focused on decision processes with prediction as an objective, and the most widely used algorithmic approach has been data-driven. In addition, the taxonomy was used to develop the COVID-19 Crowd Management project to test its usefulness as a design framework. In this regard, TDMIoT narrowed the search for decision models, validating its effectiveness in selecting an algorithmic approach for a given objective.

Development and maturity of co-word thematic clusters: the field of linked data

PurposeThis research aimed to visualize and analyze the co-word network and thematic clusters of the intellectual structure in the field of linked data during 1900–2021.Design/methodology/approachThis applied research employed a descriptive and analytical method, scientometric indicators, co-word techniques, and social network analysis. VOSviewer, SPSS, Python programming, and UCINet software were used for data analysis and network structure visualization.FindingsThe top ranks of the Web of Science (WOS) subject categorization belonged to various fields of computer science. Besides, the USA was the most prolific country. The keyword ontology had the highest frequency of co-occurrence. Ontology and semantic were the most frequent co-word pairs. In terms of the network structure, nine major topic clusters were identified based on co-occurrence, and 29 thematic clusters were identified based on hierarchical clustering. Comparisons between the two clustering techniques indicated that three clusters, namely semantic bioinformatics, knowledge representation, and semantic tools were in common. The most mature and mainstream thematic clusters were natural language processing techniques to boost modeling and visualization, context-aware knowledge discovery, probabilistic latent semantic analysis (PLSA), semantic tools, latent semantic indexing, web ontology language (OWL) syntax, and ontology-based deep learning.Originality/valueThis study adopted various techniques such as co-word analysis, social network analysis network structure visualization, and hierarchical clustering to represent a suitable, visual, methodical, and comprehensive perspective into linked data.

Fuzzy Networks for Modeling Shared Semantic Knowledge

Abstract Shared conceptualization, in the sense we take it here, is as recent a notion as the Semantic Web, but its relevance for a large variety of fields requires efficient methods of extraction and representation for both quantitative and qualitative data. This notion is particularly relevant for the investigation into, and construction of, semantic structures such as knowledge bases and taxonomies, but given the required large, often inaccurate, corpora available for search we can get only approximations. We see fuzzy description logic as an adequate medium for the representation of human semantic knowledge and propose a means to couple it with fuzzy semantic networks via the propositional Łukasiewicz fuzzy logic such that these suffice for decidability for queries over a semantic-knowledge base such as “to what degree of sharedness does it entail the instantiation C(a) for some concept C” or “what are the roles R that connect the individuals a and b to degree of sharedness ε.”

TransZero++: Cross Attribute-Guided Transformer for Zero-Shot Learning.

Zero-shot learning (ZSL) tackles the novel class recognition problem by transferring semantic knowledge from seen classes to unseen ones. Semantic knowledge is typically represented by attribute descriptions shared between different classes, which act as strong priors for localizing object attributes that represent discriminative region features, enabling significant and sufficient visual-semantic interaction for advancing ZSL. Existing attention-based models have struggled to learn inferior region features in a single image by solely using unidirectional attention, which ignore the transferable and discriminative attribute localization of visual features for representing the key semantic knowledge for effective knowledge transfer in ZSL. In this paper, we propose a cross attribute-guided Transformer network, termed TransZero++, to refine visual features and learn accurate attribute localization for key semantic knowledge representations in ZSL. Specifically, TransZero++ employs an attribute → visual Transformer sub-net (AVT) and a visual → attribute Transformer sub-net (VAT) to learn attribute-based visual features and visual-based attribute features, respectively. By further introducing feature-level and prediction-level semantical collaborative losses, the two attribute-guided transformers teach each other to learn semantic-augmented visual embeddings for key semantic knowledge representations via semantical collaborative learning. Finally, the semantic-augmented visual embeddings learned by AVT and VAT are fused to conduct desirable visual-semantic interaction cooperated with class semantic vectors for ZSL classification. Extensive experiments show that TransZero++ achieves the new state-of-the-art results on three golden ZSL benchmarks and on the large-scale ImageNet dataset. The project website is available at: https://shiming-chen.github.io/TransZero-pp/TransZero-pp.html.

(Digital Presentation) A Battery Interface Ontology for Data Interoperability and Semantic Knowledge Representation

Data drives battery development. Despite the unprecedented volume of data being generated across both industry and research today, the battery data landscape remains a patchwork of heterogeneous sources with inconsistent metadata needed to correctly interpret its meaning. A new approach is needed to ensure that reported battery data is interoperable, reusable, and machine-readable. Furthermore, there is a growing need to share knowledge about batteries among groups from different backgrounds and organizations (e.g., experimentalists, modellers, engineers, technicians, etc.).The Battery Interface Ontology (BattINFO) is designed to meet these needs [1]. An ontology is a data model that represents domain knowledge as a map of concepts and the relations between them. This allows data to be mapped to common vocabulary terms, such that any two pieces of data mapped to the same term are also mapped to each other. Additionally, the network of concepts and relations allow expert knowledge to be expressed as graphs in a machine-readable form, which can then be queried to identify links between individual pieces of data and quickly share knowledge across organizations.BattINFO is a free, open-source domain ontology for electrochemistry and batteries that facilitates battery data interoperability and enables machine-readable semantic knowledge representation. A conceptual overview is shown in Figure 1. BattINFO was developed within the scope of the EU H2020 project BIG-MAP to facilitate data interoperability between more than 30 battery research institutes and companies in Europe. This contribution introduces BattINFO, demonstrates specific use cases for how it can be applied to accelerate battery data handling and development, and presents plans for future development and integration into the global battery data landscape.[1] S. Clark et al., “Toward a Unified Description of Battery Data,” Adv. Energy Mater., vol. 2102702, 2021. Figure 1

Knowledge-Infused Text Classification for the Biomedical Domain

Extracting knowledge from unstructured text and then classifying it is gaining importance after the data explosion on the web. The traditional text classification approaches are becoming ubiquitous, but the hybrid of semantic knowledge representation with statistical techniques can be more promising. The developed method attempts to fabricate neural networks to expedite and improve the simulation of ontology-based classification. This paper weighs upon the accurate results between the ontology-based text classification and traditional classification based on the artificial neural network (ANN) using distinguished parameters such as accuracy, precision, etc. The experimental analysis shows that the proposed findings are substantially better than the conventional text classification, taking the course of action into account. The authors also ran tests to compare the results of the proposed research model with one of the latest researches, resulting in a cut above accuracy and F1 score of the proposed model for various experiments performed at the different number of hidden layers and neurons.

Comparison of biomedical relationship extraction methods and models for knowledge graph creation

Biomedical research is growing at such an exponential pace that scientists, researchers, and practitioners are no more able to cope with the amount of published literature in the domain. The knowledge presented in the literature needs to be systematized in such a way that claims and hypotheses can be easily found, accessed, and validated. Knowledge graphs can provide such a framework for semantic knowledge representation from literature. However, in order to build a knowledge graph, it is necessary to extract knowledge as relationships between biomedical entities and normalize both entities and relationship types. In this paper, we present and compare a few rule-based and machine learning-based (Naive Bayes, Random Forests as examples of traditional machine learning methods and DistilBERT, PubMedBERT, T5, and SciFive-based models as examples of modern deep learning transformers) methods for scalable relationship extraction from biomedical literature, and for the integration into the knowledge graphs. We examine how resilient are these various methods to unbalanced and fairly small datasets. Our experiments show that transformer-based models handle well both small (due to pre-training on a large dataset) and unbalanced datasets. The best performing model was the PubMedBERT-based model fine-tuned on balanced data, with a reported F1-score of 0.92. The distilBERT-based model followed with an F1-score of 0.89, performing faster and with lower resource requirements. BERT-based models performed better than T5-based generative models.

NEURAL NETWORK “PERSEPTRON” AND A SEMANTIC MODEL OF KNOWLEDGE REPRESENTATION FOR STUDYING THE DYNAMICS OF THE RF GDP

Theoretical foundations of GDP dynamics under conditions of market uncertainty and the formation of a digital economy are studied. Forecasting GDP is important because it allows to ensure balance, sustainability in the development of various sectors of the national economy, and its economic security, which determines the practical significance of the study.
 Purpose. To put forward and prove the hypothesis that with the help of the neural network model “perceptron” and the semantic model of knowledge representation, it is possible to obtain a forecast of Russia’s GDP for the next year.
 The scientific novelty lies in the fact that forecasts of the RF GDP were generated using a semantic knowledge representation model and the Perceptron AI system.
 Method or methodology of the work. The work used such methods as monographic, analytical, artificial intelligence system - perceptron, Mind histogram, as well as comparative analysis. Calculations were performed in tables XL perceptron was formed on the Deductor platform. We used a semantic knowledge representation model developed in the DOT language using the Graphviz 2.38 program.
 Results. Theoretical foundations of the use of neural networks in economic forecasts have been studied. A semantic model of knowledge representation has been formed regarding forecasting the GDP of the Russian Federation by a neural network. The developed AI-system “perceptron” is presented, designed to forecast Russia’s GDP based on input parameters representing a set of data reflecting the development of the real sector of the economy and the financial sector of Russia for 2011–2021.
 The scope of the results: economics, the financial sector, forecasting and planning of economic and financial activities.

A web-based application to support the interaction of spatial and semantic representation of knowledge

Abstract. The visualization of semantic information provides new ways for knowledge acquisition and access, since it enables users to explore and understand the meaning of information, as well as to identify associations among data and connections to existing knowledge. Maps on the other hand are extensively used as symbolic representations of spatial reality. Lately, researchers have delved deeper into formalizing the connection between the map and the semantics it conveys using knowledge-based formalisms such as knowledge graphs, ontologies, and linked data. Semantic visualization techniques may also enrich maps with the knowledge immanent therein and support knowledge visualization and exploration.The present paper describes the development of a web-based prototypical application that contains a spatial and a semantic component that interact. In addition to the representation of geospatial entities as cartographic features, the graph-based representation of semantic information is also implemented, to enable users to explore and understand the meaning and semantic relations between concepts and entities that appear on the map.

A Detection Method for Collapsed Buildings Combining Post-Earthquake High-Resolution Optical and Synthetic Aperture Radar Images

The detection of collapsed buildings based on post-earthquake remote sensing images is conducive to eliminating the dependence on pre-earthquake data, which is of great significance to carry out emergency response in time. The difficulties in obtaining or lack of elevation information, as strong evidence to determine whether buildings collapse or not, is the main challenge in the practical application of this method. On the one hand, the introduction of double bounce features in synthetic aperture radar (SAR) images are helpful to judge whether buildings collapse or not. On the other hand, because SAR images are limited by imaging mechanisms, it is necessary to introduce spatial details in optical images as supplements in the detection of collapsed buildings. Therefore, a detection method for collapsed buildings combining post-earthquake high-resolution optical and SAR images was proposed by mining complementary information between traditional visual features and double bounce features from multi-source data. In this method, a strategy of optical and SAR object set extraction based on an inscribed center (OpticalandSAR-ObjectsExtraction) was first put forward to extract a unified optical-SAR object set. Based on this, a quantitative representation of collapse semantic knowledge in double bounce (DoubleBounceCollapseSemantic) was designed to bridge a semantic gap between double bounce and collapse features of buildings. Ultimately, the final detection results were obtained based on the improved active learning support vector machines (SVMs). The multi-group experimental results of post-earthquake multi-source images show that the overall accuracy (OA) and the detection accuracy for collapsed buildings (Pcb) of the proposed method can reach more than 82.39% and 75.47%. Therefore, the proposed method is significantly superior to many advanced methods for comparison.

Structure-based knowledge acquisition from electronic lab notebooks for research data provenance documentation

BackgroundElectronic Laboratory Notebooks (ELNs) are used to document experiments and investigations in the wet-lab. Protocols in ELNs contain a detailed description of the conducted steps including the necessary information to understand the procedure and the raised research data as well as to reproduce the research investigation. The purpose of this study is to investigate whether such ELN protocols can be used to create semantic documentation of the provenance of research data by the use of ontologies and linked data methodologies.MethodsBased on an ELN protocol of a biomedical wet-lab experiment, a retrospective provenance model of the raised research data describing the details of the experiment in a machine-interpretable way is manually engineered. Furthermore, an automated approach for knowledge acquisition from ELN protocols is derived from these results. This structure-based approach exploits the structure in the experiment’s description such as headings, tables, and links, to translate the ELN protocol into a semantic knowledge representation. To satisfy the Findable, Accessible, Interoperable, and Reuseable (FAIR) guiding principles, a ready-to-publish bundle is created that contains the research data together with their semantic documentation.ResultsWhile the manual modelling efforts serve as proof of concept by employing one protocol, the automated structure-based approach demonstrates the potential generalisation with seven ELN protocols. For each of those protocols, a ready-to-publish bundle is created and, by employing the SPARQL query language, it is illustrated that questions about the processes and the obtained research data can be answered.ConclusionsThe semantic documentation of research data obtained from the ELN protocols allows for the representation of the retrospective provenance of research data in a machine-interpretable way. Research Object Crate (RO-Crate) bundles including these models enable researchers to easily share the research data including the corresponding documentation, but also to search and relate the experiment to each other.