Semantic Framework Research Articles

Isogeometric Size Optimization Design Based on Parameterized Volume Parametric Models

Traditional structural optimization design methods are based on the finite element analysis(FEA), which makes it difficult to construct a direct relationship between the design parameters and the design objective parameters in the structural design process. The FEA method needs to convert the models back and forth between the design model and the analysis or optimization model during the design process. It is a cumbersome and time-consuming work and also affects the analysis accuracy. We propose an integrated design method that seamlessly integrates process of design, simulation and optimization based on uniformity of design models, analysis models and optimization models by benefiting the advantages of volume parameterization and isogeometric analysis(IGA). The size parameters are input as high-level parameters, then the middle parameters are obtained through hierarchical mapping. Based on these parameters, the semantic feature framework composes of feature points, feature curves and feature surfaces and even feature volume is gradually constructed. By extracting paths and sections, the geometric feature framework is generated. The paths and sections are segmented to form the volume parametric sub-patches through volume parametric mapping. These sub-patches are merged into a whole volume parametric model that can be used for IGA and size driven deformation. Based on volume parametric model, a mathematical relationship is constructed between the design objective parameters and the size design parameters. Through the mathematical relationship, the sensitivity equations are derived for sensitivity analysis. Finally, an isogeometric size optimization process is complete. Thus, an integration of design process including geometric modeling, performance analysis, and structural optimization is achieved. Taking the maximum stiffness and the minimum stress as the size optimization objectives, the integrated design examples fall into four groups including single size optimization, multi sizes non-coupled optimization, multi sizes coupled optimization, and complex mechanical structure optimization. The optimization results prove that our method is effective, and it can be applied on complex mechanical parts. The designed results do not require reconstruction, thus achieving the integrated and optimized design of mechanical structures.

Korean Pansori Vocal Note Transcription Using Attention-Based Segmentation and Viterbi Decoding

In this paper, first, we delved into the experiment by comparing various attention mechanisms in the semantic pixel-wise segmentation framework to perform frame-level transcription tasks. Second, the Viterbi algorithm was utilized by transferring the knowledge of the frame-level transcription model to obtain the vocal notes of Korean Pansori. We considered a semantic pixel-wise segmentation framework for frame-level transcription as the source task and a Viterbi algorithm-based Korean Pansori note-level transcription as the target task. The primary goal of this paper was to transcribe the vocal notes of Pansori music, a traditional Korean art form. To achieve this goal, the initial step involved conducting the experiments with the source task, where a trained model was employed for vocal melody extraction. To achieve the desired vocal note transcription for the target task, the Viterbi algorithm was utilized with the frame-level transcription model. By leveraging this approach, we sought to accurately transcribe the vocal notes present in Pansori performances. The effectiveness of our attention-based segmentation methods for frame-level transcription in the source task has been compared with various algorithms using the vocal melody task of the MedleyDB dataset, enabling us to measure the voicing recall, voicing false alarm, raw pitch accuracy, raw chroma accuracy, and overall accuracy. The results of our experiments highlight the significance of attention mechanisms for enhancing the performance of frame-level music transcription models. We also conducted a visual and subjective comparison to evaluate the results of the target task for vocal note transcription. Since there was no ground truth vocal note for Pansori, this analysis provides valuable insights into the preservation and appreciation of this culturally rich art form.

Enhancing the Accessibility of Text Data in Decision Making for Capability-based Planning Using Ontology: A Perspective of Semantic Compliance

Project introduction texts provide decision-makers with essential information for project portfolio planning. These free texts offer valuable information about how developing capabilities align with their strategic goals. Capability-based Planning (CBP) process is concerned with optimizing the project portfolio to realize the planning, engineering, and delivery of these capabilities. However, up to now, the research on enhancing accessibility to free text data in the CBP decision-making process through semantic modeling is limited, leading the CBP’s decision-makers to ignore the potential advantages of existing semantic modeling methods when dealing with many free texts. This paper aims to address this gap by introducing knowl­edge modeling and mining of project introduction text corpus, leveraging semantic technology to support CBP. First, we design an ontology of capability to describe the core concepts relevant to the CBP process. Subsequently, a semantic framework based on the RDF knowl­edge graph is proposed, enabling humans and machines to comprehend project description texts. To capture the semantic data essential for CBP, a motif structure is employed to model semantic expressions, ensuring their consistency with CBP concepts through compliance checks. Finally, the effectiveness of the proposed semantic framework is evaluated by querying the project’s knowl­edge graph after semantic normalization, providing an assessment of its potential in CBP applications.

A Weakly Supervised Semantic Segmentation Framework for Medium-Resolution Forest Classification With Noisy Labels and GF-1 WFV Images

A Weakly Supervised Semantic Segmentation Framework for Medium-Resolution Forest Classification With Noisy Labels and GF-1 WFV Images

A comprehensive analysis method of urban rail transit operation accidents and safety management strategies based on text big data

Although the occurrence of urban rail transit operation accidents threatens the safety of people's lives and property, there is little research on how to systematically mine the knowledge behind accident data in China, and there is also lack of methods to support the safety management strategy of urban rail transit operation. To address these shortcomings, this study proposes a comprehensive analysis method that combines qualitative and quantitative analysis of urban rail operation accident factors and safety management strategies based on accident data, which can provide reference for systematic analysis of such accidents in the future and will help urban rail safety researchers and decision makers to take effective safety management measures. The new contributions of the method are as follows: (1) a semantic network framework for accidents and a conceptual model of accident risk control chain are proposed; (2) a method for acquiring and constructing key contents based on semantic analysis is elaborated, and indicators for evaluating the influence degree of key item and the influence possibility of entries are proposed; (3) a management matrix is established to serve safety decision making based on indicators for the degree of attention that key item of accidents should receive. In order to explain the application of the comprehensive analysis method, 910 urban rail transit operation safety accident cases are analyzed in detail, and suggestions for the actual operation management of urban rail transit are proposed in the light of the actual situation of urban rail transit operation in China.

A Semantic Controllable Long Text Steganography Framework Based on LLM Prompt Engineering and Knowledge Graph

A Semantic Controllable Long Text Steganography Framework Based on LLM Prompt Engineering and Knowledge Graph

Semantic agent framework for automated flood assessment using dynamic knowledge graphs

Abstract This article proposes a framework of linked software agents that continuously interact with an underlying knowledge graph to automatically assess the impacts of potential flooding events. It builds on the idea of connected digital twins based on the World Avatar dynamic knowledge graph to create a semantically rich asset of data, knowledge, and computational capabilities accessible to humans, applications, and artificial intelligence. We develop three new ontologies to describe and link environmental measurements and their respective reporting stations, flood events, and their potential impact on population and built infrastructure as well as the built environment of a city itself. These coupled ontologies are deployed to dynamically instantiate near real-time data from multiple fragmented sources into the World Avatar. Sequences of autonomous agents connected via the derived information framework automatically assess consequences of newly instantiated data, such as newly raised flood warnings, and cascade respective updates through the graph to ensure up-to-date insights into the number of people and building stock value at risk. Although we showcase the strength of this technology in the context of flooding, our findings suggest that this system-of-systems approach is a promising solution to build holistic digital twins for various other contexts and use cases to support truly interoperable and smart cities.

Semantic search framework over knowledge bases using embeddings-based similarity

With tremendous progress attained towards AI, understanding context from user queries has become a focus area to provide precise answers. In this direction, recent machine learning-based approaches are trying to extract semantic features from queries. Being keywords driven, these approaches underperform to cater semantics in the queries. To overcome this reliance on syntactic representations, semantics-based methods utilizing Knowledge Bases are being augmented to the query processing systems. In this paper, a novel semantic search framework over Knowledge Bases has been proposed and implemented. Semantic similarity between query and predicates is computed using BERTbased embeddings. Additionally, a dataset of semantically similar sentence variations is generated using ChatGPT for analysing the accuracy of the implemented system over YAGO Knowledge Base.

HiSIF-DTA: A Hierarchical Semantic Information Fusion Framework for Drug-Target Affinity Prediction.

Accurately identifying drug-target affinity (DTA) plays a significant role in promoting drug discovery and has attracted increasing attention in recent years. Exploring appropriate protein representation methods and increasing the abundance of protein information is critical in enhancing the accuracy of DTA prediction. Recently, numerous deep learning-based models have been proposed to utilize the sequential or structural features of target proteins. However, these models capture only the low-order semantics that exist in a single protein, while the high-order semantics abundant in biological networks are largely ignored. In this article, we propose HiSIF-DTA'a hierarchical semantic information fusion framework for DTA prediction. In this framework, a hierarchical protein graph is constructed that includes not only contact maps as low-order structural semantics but also protein-rotein interaction (PPI) networks as high-order functional semantics. Particularly, two distinct hierarchical fusion strategies (i.e., Top-down and Bottom-Up) are designed to integrate the different protein semantics, therefore contributing to a richer protein representation. Comprehensive experimental results demonstrate that HiSIF-DTA outperforms current state-of-the-art methods for prediction on the benchmark datasets of the DTA task. Further validation on binary tasks and visualization analysis demonstrates the generalization and interpretation abilities of the proposed method.

A Multi-Task Oriented Semantic Communication Framework for Autonomous Vehicles

A Multi-Task Oriented Semantic Communication Framework for Autonomous Vehicles

Artificial Intelligence Techniques in Semantic Web Services

It relates to the conventional machine interaction services, is the server end of a Client-Server system relates to the data on web that is understandable by the machines, its primary objective is to permit the humans and machines together. The World Wide Web is progressing quickly in the direction of the concept of semantic language for database. The ontology tool Protégé consists of the SPARQL Query Tab to execute the queries, whereas the alternative is the Apache Jena Fuseki. The semantic web framework is comprised of different Application interfaces that interacting collectively to execute RDF data. The Fuseki supports the SPARQL queries over Hyper Text Transfer Protocol. Apache Jena Fuseki is based framework for developing Artificial Intelligence Techniques in Semantic Web Services. The proposed methodology can handle queries of various convolutions and complexities. Semantic web services on the server end provide structured and standardized information about the services they offer, typically using technologies related to ontology ensure that the data is both readable and process by machines based on ontology. The Experimental Analysis is given along with the results based on different ontology based technologies. In the disciplines of design, Thinking and implementation, along with representation ontologies are used for semantic web services.

Semantic analysis and construction of English discourse based on neural network

Abstract Accurate recognition and analysis of English semantics is of great significance in the application field of computer and artificial intelligence technology. Traditional text analysis technology has poor real-time analysis accuracy and efficiency. This paper constructs an English semantic analysis model based on neural networks. By designing an English semantic framework that integrates LSTM and RNN and optimizing it on the basis of LSTM and RNN models, an English semantic analysis model with improved attention mechanisms and Bi-LSTM is constructed. By comparing the Bi-LSTM+Attention model proposed in this paper with other semantic analysis models, the classification accuracy of the model constructed in this paper reaches 98%. Through the practical application of the model to analyze the semantic coherence of English, the average score of the semantic coherence quality measured by the model is 15.76, and the average score of the manual judgment is 16.03, which indicates that the application of the semantic analysis model constructed in this paper is close to the effect of the manual judgment.

The Conceptual Relationships in N+N Compounds in Arabic Compared to English

The current study investigates the semantic or conceptual relations that exist between two compound elements in Arabic based on the theoretical framework of Conceptual Semantics and a modified version of Parallel Architecture called Relational Morphology (RM). It also explores the similarities and/or differences between these relations and those identified by Jackendoff pertaining to English compounds. To this end, a corpus of Arabic compounds was collected and analysed to identify the type of conceptual basis on which both lexicalised and novel compounds are formed in Arabic. The analysis revealed that all conceptual relations found between elements of English compounds can also be found in Arabic, yet the direction of the head was different in the latter. The analysis also demonstrated that Arabic compounds exhibit other types of conceptual relations which were not mentioned by Jackendoff. It is suggested that both regular stored lexicalised compounds that have idiosyncratic interpretations and novel compounds (with the majority being formed based on Synthetic Genitive Construction) can be reconciled under the approach of Relational Morphology, as they can be generated by the same compound schema.

Developing a semantic framework for categorizing IoT agriculture sensor data: A machine learning and web semantics approach

This study introduces a Semantic Framework for Categorizing IoT Agriculture Sensor Data, leveraging Machine Learning and Web Semantics. IoT sensors in agriculture generate vast real-time data on crucial factors like soil conditions and weather, promising optimization in resource use and crop yields. While machine learning aids data categorization, semantic aspects often remain unexplored. By combining machine learning with web semantics (RDF and OWL), this research establishes a structured framework that not only categorizes data but also links it to actionable farming recommendations. Methodologically, it involves data collection, preprocessing, machine learning, and semantic integration. Performance evaluation through metrics and visualizations reveals the framework's effectiveness, aiding decision-making in precision agriculture. This study contributes to IoT-based precision agriculture by bridging the gap between raw sensor data and actionable insights, empowering a semantic framework for contextual categorization and recommendation generation. The fusion of machine learning and web semantics holds transformative potential for agriculture, enhancing data management and decision-making processes.

Autonomous detection of steel corrosion spatial variability in reinforced concrete using X-ray technology and deep learning-based semantic segmentation

Correctly determining the spatial distribution of steel corrosion within a structural member is critical for estimating the remaining service life of deteriorating reinforced concrete (RC) structures. While X-ray technology serves as a nondestructive inspection method, existing challenges persist, particularly in semi-automated corrosion boundary detection. This paper describes a deep learning-based semantic segmentation framework to autonomously detect X-ray images associated with RC, facilitating the visualization of nonuniform steel corrosion distribution. X-ray images were collected from a comprehensive experiment using RC specimens with various structural details by two accelerated corrosion methods. Four deep learning models were constructed, trained, and compared based on the database containing the original X-ray images and the corresponding pixel-level labels. The results demonstrate that the proposed autonomous detection method can segment uncorroded steel at a very high level of global accuracy without time-consuming work, outperforming traditional methods in terms of both accuracy and efficiency.

Development of Interoperability Principles for Disaster and Emergency Management System of Türkiye

Due to their complex nature, disasters and emergencies require a data-intensive management system in which many actors from different sectors participate and simultaneous processes are managed. Due to this complexity, there are problems in managing and sharing process services and geographic data effectively. The key to solving these problems is the introduction of a complete interoperability model. In this context, interoperability models based on existing international standards are analyzed in this study. FEI and EIF models are taken as basis in terms of disaster and emergency management interoperability requirements. Accordingly, the interoperability reference model was created in legal, organizational, semantic and technical frameworks to meet the interoperability levels. This general framework model has been evaluated in the service and data layers that need to be fully defined in disaster and emergency management phases, and a basic model including interoperability solutions for the service and data model has been created.

Development of Semantic-Web Based Forest Resources Database

There are some forest resources (such as Trees, Shrubs, and Water bodies) in the university main campus. These resources are used for various purposes ranging from teaching, researches medicinal and domestic purposes and among many others. However, there is no record of official documentation on them. Thus, this study aims to provide a semantic-web based database system for efficient management of forest resources. We adopted the W3C semantic web framework in the design and development of this work. Specifically, we focused on the layers 1 to 5 of the framework. The layers include; Resource Description Framework (RDF), Web Ontology Language (OWL), Extensible Markup Language (XML), User Interface). The Uschold’s and Kings’ method was applied in creating the ontology. The implementation process was carried out by populating the database with the Taxonomies of the Trees on University of Energy and Natural Resources (U.E.N.R) campus by means of creating an Ontology File using protégé. Some of the basic information that we provided by the proposed semantic web database include trees, their scientific name, geographical coordinate, physical location, medicinal value and other important data. We performed alpha test by involving six senior forest science students who used the application, evaluated it and came out with a good result which is shown in the chapter four of this documentation.

DeepGraFT: A novel semantic segmentation auxiliary ROI-based deep learning framework for effective fundus tessellation classification

Fundus tessellation (FT) is a prevalent clinical feature associated with myopia and has implications in the development of myopic maculopathy, which causes irreversible visual impairment. Accurate classification of FT in color fundus photo can help predict the disease progression and prognosis. However, the lack of precise detection and classification tools has created an unmet medical need, underscoring the importance of exploring the clinical utility of FT. Thus, to address this gap, we introduce an automatic FT grading system (called DeepGraFT) using classification-and-segmentation co-decision models by deep learning. ConvNeXt, utilizing transfer learning from pretrained ImageNet weights, was employed for the classification algorithm, aligning with a region of interest based on the ETDRS grading system to boost performance. A segmentation model was developed to detect FT exits, complementing the classification for improved grading accuracy. The training set of DeepGraFT was from our in-house cohort (MAGIC), and the validation sets consisted of the rest part of in-house cohort and an independent public cohort (UK Biobank). DeepGraFT demonstrated a high performance in the training stage and achieved an impressive accuracy in validation phase (in-house cohort: 86.85 %; public cohort: 81.50 %). Furthermore, our findings demonstrated that DeepGraFT surpasses machine learning-based classification models in FT classification, achieving a 5.57 % increase in accuracy. Ablation analysis revealed that the introduced modules significantly enhanced classification effectiveness and elevated accuracy from 79.85 % to 86.85 %. Further analysis using the results provided by DeepGraFT unveiled a significant negative association between FT and spherical equivalent (SE) in the UK Biobank cohort. In conclusion, DeepGraFT accentuates potential benefits of the deep learning model in automating the grading of FT and allows for potential utility as a clinical-decision support tool for predicting progression of pathological myopia.

Dynamic semantic structure distillation for low-resolution fine-grained recognition

Low-resolution images are ubiquitous in real applications such as surveillance and mobile photography. However, existing fine-grained approaches usually suffer catastrophic failures when dealing with low-resolution inputs. This is because their learning strategy inherently depends on the semantic structure of the pre-trained model, resulting in poor robustness and generalization. To mitigate this limitation, we propose a dynamic semantic structure distillation learning framework. Our method first facilitates knowledge distillation of diverse semantic structures by perturbing the composition of semantic components and then utilizes a decoupled distillation objective to prevent the loss of primary semantic part relation knowledge. We evaluate our proposed approach on two knowledge distillation tasks: high-to-low resolution and large-to-small model. The experimental results show that our proposed approach significantly outperforms existing methods in low-resolution fine-grained image classification tasks. This indicates that it can effectively distill knowledge from high-resolution teacher models to low-resolution student models. Furthermore, we demonstrate the effectiveness of our approach in general image classification and standard knowledge distillation tasks.

On the semantics of (negated) approximative kaada in Classical Arabic: a case for embedded exhaustification

Abstract The purpose of this paper is twofold. First, it proposes a compositional semantic analysis for approximative kaada in Classical Arabic which has the property of being a clause-level rather than predicate-level operator: [KAADA α] is paraphrased as [α is false but there exists β close to α such that β is true]. The analysis is based on the integrated semantics of Penka (Penka, Doris. 2006. Almost there: The meaning of almost. Proceedings of Sinn und Bedeutung 10. 275–286) in which the negativity component of kaada is interpreted as part of its lexical meaning within a situation semantics framework. It also makes a novel use of the concept of semantic closeness among situations in terms of the intuitively appealing dimension of the relative completion of an event. Second, the paper presents and analyzes an instance of a puzzle noted by most descriptive Classical Arabic grammarians (see, for example, al-Ahmad, Ayman. 2021. Dalalit KAADA l-manfiah fi l-Qurʾaan l-kareem. Journal of the Association of Arab Universities for Literature 9(2). 838–859) in which negated kaada has an interpretation where the negativity implication disappears altogether. The paper offers a solution to the puzzle based on a previous suggestion made by Kilbourn-Ceron (Kilbourn-Ceron, Oriana. 2016. Embedded exhaustification: Evidence from almost. Journal of Semantics 34(1). 43–60) in which an instance of embedded exhaustification appears to apply in approximatives in certain contexts in such a way that the negativity inference is obscured.