Ontological Knowledge Research Articles

Supporting vision-language model few-shot inference with confounder-pruned knowledge prompt.

Vision-language models are pre-trained by aligning image-text pairs in a common space to deal with open-set visual concepts. Recent works adopt fixed or learnable prompts, i.e., classification weights are synthesized from natural language descriptions of task-relevant categories, to reduce the gap between tasks during the pre-training and inference phases. However, how and what prompts can improve inference performance remains unclear. In this paper, we explicitly clarify the importance of incorporating semantic information into prompts, while existing prompting methods generate prompts without sufficiently exploring the semantic information of textual labels. Manually constructing prompts with rich semantics requires domain expertise and is extremely time-consuming. To cope with this issue, we propose a knowledge-aware prompt learning method, namely Confounder-pruned Knowledge Prompt (CPKP), which retrieves an ontology knowledge graph by treating the textual label as a query to extract task-relevant semantic information. CPKP further introduces a double-tier confounder-pruning procedure to refine the derived semantic information. Adhering to the individual causal effect principle, the graph-tier confounders are gradually identified and phased out. The feature-tier confounders are eliminated by following the maximum entropy principle in information theory. Empirically, the evaluations demonstrate the effectiveness of CPKP in few-shot inference, e.g., with only two shots, CPKP outperforms the manual-prompt method by 4.64% and the learnable-prompt method by 1.09% on average.

Interweaving academic insights: advancing university knowledge management through a strategic data fabric framework

PurposeEffective knowledge management in large academic institutions is crucial for fostering innovation and improving educational practices. However, these institutions often face challenges, such as data fragmentation, siloed information systems and the complexity of integrating different data sources from various departments with complex hierarchical structures. To address these problems, the authors proposed a data fabric strategic framework that improves and enhances knowledge management by leveraging ontologies and knowledge graphs. This study aims to investigate the potential of knowledge graphs, ontological knowledge modelling and knowledge representation to improve knowledge management in large academic institutions. It also describes how technology can enhance knowledge accessibility and exchanges and improve decision-making processes based on insights from complex educational systems.Design/methodology/approachThis study uses coordination theory as a foundational framework to analyse intricate data systems in preparation for constructing, the Wizard of Oz method to facilitate the systematic organisation and management of information and the execution of an ontology-based data fabric framework and knowledge graphs. The authors propose a data fabric strategic framework aimed at improving knowledge management by leveraging ontologies and knowledge graphs.FindingsThe final evaluation demonstrates that this approach effectively breaks down data silos, promotes research collaboration and improves decision-making processes in large academic settings, offering solution-oriented data fabric technologies applicable to universities and university federations globally.Practical implicationsThe proposed system provides a more efficient way of managing and connecting fragmented academic resources, improving accessibility for both learners and educators. By interconnecting and streaming knowledge management process, the system can reduce not only operational costs but also expenses on doing scientific research.Originality/valueAcademic institutions prioritise time efficiency when acquiring vital data for improved scientific results. This emphasis extends beyond data governance to focus on how collective intelligence might improve organisational performance. The academic community has enhanced data utilisation through the implementation of data fabric technologies to improve data accessibility and data line tracking.

SciAgents: Automating Scientific Discovery Through Bioinspired Multi-Agent Intelligent Graph Reasoning.

A key challenge in artificial intelligence (AI) is the creation of systems capable of autonomously advancing scientific understanding by exploring novel domains, identifying complex patterns, and uncovering previously unseen connections in vast scientific data. In this work, SciAgents, an approach that leverages three core concepts is presented: (1) large-scale ontological knowledge graphs to organize and interconnect diverse scientific concepts, (2) a suite of large language models (LLMs) and data retrieval tools, and (3) multi-agent systems with in-situ learning capabilities. Applied to biologically inspired materials, SciAgents reveals hidden interdisciplinary relationships that werepreviously considered unrelated, achieving a scale, precision, and exploratory power that surpasses human research methods. The framework autonomously generates and refines research hypotheses, elucidating underlying mechanisms, design principles, and unexpected material properties. By integrating these capabilities in a modular fashion, the system yields material discoveries, critiques and improves existing hypotheses, retrieves up-to-date data about existing research, and highlights strengths and limitations. This is achieved by harnessing a "swarm of intelligence" similar to biological systems, providing new avenues for discovery. How this model accelerates the development of advanced materials by unlocking Nature's design principles, resulting in a new biocomposite with enhanced mechanical properties and improved sustainability through energy-efficient production is shown.

Cognitive digital twin in manufacturing process: integrating the knowledge graph for enhanced human-centric Industry 5.0

Industry 5.0 emphasises human-centric intelligent manufacturing, posing challenges in integrating human expertise with advanced machine capabilities. To address these challenges, a novel three-layer cognitive digital twin model based on knowledge graphs is proposed, designed to integrate workers’ knowledge and experience into intelligent manufacturing processes. This model comprises three layers: an ontology layer that constructs a foundational process knowledge ontology library; a knowledge layer that maps real-time data to dynamically update digital models; and a cognitive layer that utilises machine learning, knowledge reasoning, and knowledge mining for advanced analysis, state understanding, and model evolution. The model promotes user interaction through intuitive interfaces and a Q&A system, leveraging techniques such as knowledge reasoning and querying to support decision-making and enhance worker engagement. Validated through a system implemented for aero-engine blade production, this cognitive digital twin model leverages human expertise and machine capabilities to enhance process control, quality management, and overall efficiency. The proposed approach demonstrates significant potential for advancing personalised human-machine interaction in manufacturing, truly embodying the value of a human-centric approach and paving the way for future developments in the field.

Ontology and production rules-based dynamic knowledge base construction methodology for machining process

With advancements in manufacturing, knowledge engineering has become important in supporting intelligent decision-making within manufacturing systems. However, existing process knowledge bases, integral to knowledge engineering, and essential for machining efficiency, product cost, and production cycles by integrating multi-source knowledge, are limited to generality, scalability, and adaptability to real production environments. These constraints undermine the application and reliability of process knowledge bases in decision-making. To overcome these challenges, an approach to constructing a dynamic machining process knowledge base (DMPKB) utilizing ontology and production rules is proposed. Firstly, a machining process knowledge model is developed by reorganizing concepts and relations to restructure process cases and experiences, thereby building a comprehensive knowledge base. Secondly, different update strategies are devised to fulfill the requirements of various components within the knowledge base. Finally, the effectiveness is validated by constructing a DMPKB for CNC boring machine bearing seats. Meanwhile, application verification is performed by generating process plans for a CNC boring machine bearing seat, showcasing the feasibility and utility of the developed knowledge base.

Dynamic Retrieval Augmented Generation of Ontologies using Artificial Intelligence (DRAGON-AI)

BackgroundOntologies are fundamental components of informatics infrastructure in domains such as biomedical, environmental, and food sciences, representing consensus knowledge in an accurate and computable form. However, their construction and maintenance demand substantial resources and necessitate substantial collaboration between domain experts, curators, and ontology experts.We present Dynamic Retrieval Augmented Generation of Ontologies using AI (DRAGON-AI), an ontology generation method employing Large Language Models (LLMs) and Retrieval Augmented Generation (RAG). DRAGON-AI can generate textual and logical ontology components, drawing from existing knowledge in multiple ontologies and unstructured text sources.ResultsWe assessed performance of DRAGON-AI on de novo term construction across ten diverse ontologies, making use of extensive manual evaluation of results. Our method has high precision for relationship generation, but has slightly lower precision than from logic-based reasoning. Our method is also able to generate definitions deemed acceptable by expert evaluators, but these scored worse than human-authored definitions. Notably, evaluators with the highest level of confidence in a domain were better able to discern flaws in AI-generated definitions. We also demonstrated the ability of DRAGON-AI to incorporate natural language instructions in the form of GitHub issues.ConclusionsThese findings suggest DRAGON-AI's potential to substantially aid the manual ontology construction process. However, our results also underscore the importance of having expert curators and ontology editors drive the ontology generation process.

An ontology-based multi-hazard coupling accidents simulation and deduction system for underground utility tunnel - A case study of earthquake-induced disaster chain

Integrated underground utility tunnels are increasingly crucial in modern cities, addressing the pressing need for sustainable urban development. However, their extensive centralization amplifies both the complexity and scale of potential risks. When a utility tunnel accident occurs, it is possible to trigger a sequence of cascading events, thereby resulting a complex coupling accident. While previous research has predominantly focused on individual hazards, understanding multi-hazard coupling accidents presents significant challenges and lacks effective methodologies. In this paper, we propose an integrated system utilizing ontology technology and knowledge base construction for simulating and deducing coupling accidents in urban utility tunnels. Specifically, by extending ontology techniques to emergency decision-making and adopting the triangular framework for public safety, we establish a multidimensional information ontology for utility tunnel emergencies. Furthermore, a knowledge base for typical coupling accident evolution paths is established based on the event chain and contingency plan chain theory. Through integration with a multi-hazard accident basic database that serves the conditional, investigative and decision-making node within the evolution path, the simulation and deduction system is formulated, boasting a user-friendly visual interface, interactive functionality, and seamless applicability for widespread adoption. A case study demonstrates the system ability to support multiple paths and unified mapping deduction, offering practical emergency decision-making suggestions to mitigate cascading events in urban utility tunnels.

Онтологиялық модельдеу көмегімен "Ontomektep" пәнаралық байланыстарды анықтау

Ontological knowledge bases have a great advantage over relational databases. Ontologies of higher educational institutions are being developed in the Republic of Kazakhstan. Ontologies of educational process in school are relevant, but not fully developed. The purpose of the article is to create a conceptual model of the educational content of schoolchildren, its use provides high-quality management, flexible selection and systematization of school data and knowledge. To achieve this goal, we use the methods of ontological engineering, because the semantic description of knowledge about the educational process at school increases the quality of education of schoolchildren. To develop a school ontology, it is necessary to perform a sequence of activities and implement ontological engineering. First, we analyze the subject area, perform a synthesis of concepts and connections, and then select objects, features, connections, and processes. Ontology creation, or visual ontology engineering, is a powerful tool for visualizing the knowledge structures of any subject area. Currently, there are a large number of ontology editors. Each editor has its advantages and disadvantages. With the help of the created ontological model and filling the knowledge base of the school, it is possible to determine the interdisciplinary connections of mathematics and informatics. You can continue to enter the necessary data in this knowledge base and thereby update the school data.

Using Ontology to Analyze English Comments on Social Networks

Chatbots have become interesting for many users as technology becomes more and more advanced. The need for information exchange among people through computer systems is increasing daily, raising the preference for using chatbots in most countries. Since Vietnam is such a developing country with a variety of ethnic groups, it requires much attention to the proliferation of social networks and the expansion of the cooperative economy. Regarding social networks, the inappropriate use of words in everyday life has become a significant issue. There are mixed reviews of praise and criticism on social networks; and we try to reduce the negative language use and improve the quality of using social networks language. We aim to meet users’ needs on social networks, promote economic development, and address social issues more effectively. To achieve these goals, in this paper we propose a deep learning technique using ontology knowledge mining to collect and process comments on social networks. This approach aims to enhance the user experience and facilitate the exchange of information among people by mining opinions in comments. Experimental results demonstrate that our method outperforms the conventional approach.

Construction of Classroom Culture of International Chinese Teachers from the Perspective of Effective Teaching

Culture and economy are two subsystems that are indispensable and closely linked in a country's development process. Economy determines culture, and culture is a reflection of politics and economy and reacts to economy. With the continuous rapid development of China's economy, "Chinese", as a unique cultural element in China, is increasingly active and spreading on the global international stage with the development of China's economy.​ Therefore, China's language and the cultural civilization factors attached to the language are in urgent need of "going out", they need a "missionary" who not only has the Chinese ontology knowledge but also has a high degree of cultural cognition to spread Chinese characters and Chinese civilization. At this time, international Chinese teachers who have a high degree of conformity with internal and external standards and who are familiar with cultural teaching methods and teaching techniques become indispensable. It is imperative to improve the comprehensive quality of international Chinese teachers and promote the effective teaching of international Chinese classroom culture.

OGSS: An Ontology-Guided and Scheduled-Sampling Approach for Overlapping Event Extraction

Event extraction is a complex and challenging task in the field of information extraction. It aims to identify event types, triggers, and argument information from the text. In recent years, overlapping event extraction has attracted the attention of researchers because of its higher challenge and practicability, and some work has carried out in-depth research on overlapping event extraction and achieved remarkable results. But these works (1) ignore the role of ontology knowledge in event extraction; (2) use the same semantic encoding for multi-stage models, lacking consideration for the independent characteristics of extraction tasks such as event types, triggers, and arguments; and (3) face issues in the training process of multi-stage models, such as error cascading and slow convergence. To address the above issues, we propose an ontology-guided and scheduled-sampling approach for overlapping event extraction, termed as OGSS. First, we design a symmetric matrix for event ontology knowledge representation and integrate it into the semantic encoding process, infusing ontology knowledge into event extraction. Second, for extraction targets such as event types, triggers, and arguments, we process the semantic encoding according to the characteristics of each extraction target, obtaining semantic representations tailored for each subtask. Finally, we view multi-stage predictions as sequential outputs of a joint model, using a scheduled sampling strategy between subtasks to effectively mitigate the cascading propagation of errors during training and accelerate model convergence. We conduct extensive experiments on the FewFc event extraction benchmark dataset. The results show that OGSS achieves significant improvements in overlapping event extraction tasks compared to previous methods.

Alternative Tourism – an Ontology of Bulgarian National Revival Residential Architecture for the Purposes of an Intelligent Tourist Guide

In the article, we will explain the concept of alternative tourism and its subtypes, we will dwell on its role in the preservation of cultural and historical heritage of an architectural type by means of technological solutions – an intelligent guide with ontological knowledge.

Hierarchical clustering algorithm to generate ontology network

Ontologies have played a crucial role in addressing challenges related to knowledge modeling, artificial intelligence, clustering, and classification. Within the realm of web services, ontologies primarily serve to describe web service interfaces, facilitating the automation of functional processes, particularly to guide the registers in discovering services. Enriching this model with a shared ontology (knowledge base) can improve the efficiency of web services technology by guiding web services processes. This shared ontology plays the role of intermediary between queries and registries, and can be shared between different registries. However, manually constructing ontologies is a labor-intensive and expensive endeavor, necessitating domain expertise. Automatic ontology construction approaches present compelling alternatives, offering advantages in terms of both time and financial costs. In this article, we advocate for the utilization of the Hierarchical Clustering algorithm to generate an ontological network (Service Ontology (SO)) from a semantic web services corpus. Experimental validations are presented to demonstrate the effectiveness of the proposed methodology.

Accelerating scientific discovery with generative knowledge extraction, graph-based representation, and multimodal intelligent graph reasoning

Abstract Leveraging generative Artificial Intelligence (AI), we have transformed a dataset comprising 1,000 scientific papers focused on biological materials into a comprehensive ontological knowledge graph. Through an in-depth structural analysis of this graph, we have calculated node degrees, identified communities along with their connectivities, and evaluated clustering coefficients and betweenness centrality of pivotal nodes, uncovering fascinating knowledge architectures. We find that the graph has an inherently scale-free nature, shows a high level of connectedness, and can be used as a rich source for downstream graph reasoning by taking advantage of transitive and isomorphic properties to reveal insights into unprecedented interdisciplinary relationships that can be used to answer queries, identify gaps in knowledge, propose never-before-seen material designs, and predict material behaviors. Using a large language embedding model we compute deep node representations and use combinatorial node similarity ranking to develop a path sampling strategy that allows us to link dissimilar concepts that have previously not been related. One comparison revealed detailed structural parallels between biological materials and Beethoven's 9th Symphony, highlighting shared patterns of complexity through isomorphic mapping. In another example, the algorithm proposed an innovative hierarchical mycelium-based composite based on integrating path sampling with principles extracted from Kandinsky's `Composition VII' painting. The resulting material integrates an innovative set of concepts that include a balance of chaos and order, adjustable porosity, mechanical strength, and complex patterned chemical functionalization. We uncover other isomorphisms across science, technology and art, revealing a nuanced ontology of immanence that reveal a context-dependent heterarchical interplay of constituents. Because our method transcends established disciplinary boundaries through diverse data modalities (graphs, images, text, numerical data, etc.), graph-based generative AI achieves a far higher degree of novelty, explorative capacity, and technical detail, than conventional approaches and establishes a widely useful framework for innovation by revealing hidden connections.

BVTED: A Specialized Bilingual (Chinese–English) Dataset for Vulnerability Triple Extraction Tasks

Extracting knowledge from cyber threat intelligence is essential for understanding cyber threats and implementing proactive defense measures. However, there is a lack of open datasets in the Chinese cybersecurity field that support both entity and relation extraction tasks. This paper addresses this gap by analyzing vulnerability description texts, which are standardized and knowledge-dense, to create a vulnerability knowledge ontology comprising 13 entities and 15 relations. We annotated 27,311 unique vulnerability description sentences from the China National Vulnerability Database, resulting in a dataset named BVTED for cybersecurity knowledge triple extraction tasks. BVTED contains 97,391 entities and 69,614 relations, with entities expressed in a mix of Chinese and English. To evaluate the dataset’s value, we trained five deep learning-based named entity recognition models, two relation extraction models, and two joint entity–relation extraction models on BVTED. Experimental results demonstrate that models trained on this dataset achieve excellent performance in vulnerability knowledge extraction tasks. This work enhances the extraction of cybersecurity knowledge triples from mixed Chinese and English threat intelligence corpora by providing a comprehensive ontology and a new dataset, significantly aiding in the mining, analysis and utilization of the knowledge embedded in cyber threat intelligence.

Development of Composite Intelligent Services Operating in a Heterogeneous Environment

The paper discusses methods, approaches, tools and technologies for composite service development. The expediency of applying the considered principles to create software application of so-called composite artificial intelligence (AI) is noted in order to implement the latter within the "distributed artificial intelligence as a service" (DAIaaS) paradigm. With the participation of the authors, the IACPaaS (Intelligent Applications, Control and Platform as a Service) cloud platform has been created and continues to be actively developed. It is intended for the creation, use and maintenance of multi-agent intelligent services that process ontological knowledge graphs. Currently, the platform provides such cloud service delivery models as PaaS (platform as a service), SaaS (software as a service), DaaS (data as a service) and Desktop as a Service. One of development directions is to provide support for the DAIaaS paradigm. It consists in the designing and implementation of tools that make it possible to use the platform and its services as components of composite applications (including AI ones) operating in a heterogeneous distributed computing environment. As a result, language and software tools were created that allow the development of distributed intelligent systems, some of the components of which are located on the IACPaaS platform, and the rest are located elsewhere. Depending on where the leading component is placed, certain tools and technologies implemented on the platform are used. The article also contains examples of created services for several domains (medicine, transport modeling, laser additive manufacturing) and options for using the platform.

A systematic mapping study of applying artificial intelligence and data analytics in law and jurisdiction: legal ontologies based on technology

The use of technology in legal reasoning has become the mainstream direction of legalsystems and governments. Especially, higher quality services provided by artificial intelligence and bigdata technologies enable lawyers and government officials to reduce the risk of inappropriatelegislation and act efficiently. Although a significant number of researches have been conducted sofar in that area, there is still an increasing need for structuring knowledge ontologies about the use ofartificial intelligence and data analysis in law and jurisdiction. Therefore, this paper aims to provide acomprehensive overview of applications of those higher-level technologies in legal reasoning andprocessing by identifying, classifying, and evaluating existing literature related to that area of research.The main procedures of this study include planning of the study (definitions of research questions andthe scope and establishment of the search strategy and selection criteria) and conducting of the study(analysis & mapping, data extraction & classification, selection of primary studies, and searchexecution). As a result of this systematic mapping study, the findings include a classified portfolio ofpublications, higher level mapping of the structure of existing literature, and visualized status of thetarget research area about pre-established research questions. Based on the findings of thissystematic study, the author draws implications and guidelines for lawyers and government officialswho have an interest in using such first-class technologies in their jobs. Finally, this article also providesresearch agenda to fill in the gaps in this knowledge domain and spot research trends.

»Relationships are Key« A Semantic Relationship Awareness Framework for Operational Technology Monitoring

Critical infrastructures in areas like road traffic management naturally rely on the broad use of "Operational Technology (OT)" to ensure efficient and safe road traffic monitoring (RTM) through "OT objects", like sensors and actuators, whereby monitoring OT itself ("OTM") is evenly crucial. OTM is highly challenging, not least due to massive heterogeneity of OT, immense complexity and size, and omnipresence of evolution. As a consequence, knowledge about interdependencies between OT objects in form of semantic relationships is often outdated or simply not available. Thus, in case of incidents, detection of cause and effect in the sense of a situational picture of OT is missing. In order to counteract this fundamental deficiency, we aim to automatically discover semantic relationships between OT objects, to build up an ontological knowledge base as prerequisite for achieving OT situation awareness. Thereby, the contribution of this paper is threefold. First, a systematic exploration of the induced challenges is provided, derived from an in-depth analysis of real-world OT message logs in the area of RTM. Based on that, we sketch out a research roadmap, thereby guiding the identification of existing concepts and technologies appearing to be useful for realizing a framework for semantic relationship awareness, being the crucial pre-step for achieving OT situation awareness. Finally, a first proof-of-concept prototype is put forward, complemented by an evaluation of its applicability and a detailed comparison to related approaches.

Earthquake event knowledge graph construction and reasoning

Efficient decision-making in earthquake emergencies plays a crucial role in ensuring individual safety, protecting personal property, and maintaining societal stability. However, traditional approaches to earthquake emergency decision-making rely on manual analysis or rule-based methods, which often struggle to fully leverage the wealth of information and uncover hidden data connections. Consequently, the efficiency of earthquake emergency decision-making is compromised. To address this issue, this study proposes a method for constructing an earthquake event knowledge graph and utilizing it for decision-making in earthquake emergencies. Firstly, specialized earthquake event knowledge ontology is developed, tailored to the unique characteristics of earthquake event data. Secondly, structured instances of earthquake event knowledge are extracted from text using transfer learning techniques, enabling the construction of the earthquake event knowledge graph. Thirdly, the earthquake event knowledge is represented as multidimensional vectors using knowledge graph representation learning technology. This facilitates the identification of similar earthquake events through inference based on vector similarity computation. In conclusion, the results of a case-based study demonstrate the effectiveness of the proposed method in providing accurate outcomes, facilitating earthquake event matching, enabling the retrieval and reuse of historical earthquake event knowledge, and serving as a valuable reference for earthquake emergency decision-making.

Product innovation design approach driven by implicit relationship completion via patent knowledge graph

Product innovation design process involves a great deal of discrete engineering knowledge, limiting the ability of designers to quickly utilize this knowledge to support design innovation. Nowadays, innovation design based on knowledge graphs has enhanced the ability to explore design knowledge, improving the efficiency of knowledge retrieval. Previous studies have focused on mining more design knowledge to enrich the knowledge graph overlooks the implicit relationships with potential value among design knowledge, wasting design resources. To address these issues, an approach for product innovation design based on implicit knowledge relationship completion in the patent knowledge graph is proposed, which explores the implicit relationships between design knowledge to provide new knowledge satisfying design preferences and enhance the innovativeness of solutions. First, a requirements-function-structure-benefit (RFSB) knowledge ontology is constructed and extracted from the benefit knowledge of patents to build the knowledge graph. Second, an implicit relationship completion model based on the similarity of function or benefit entities explores the implicit relationships, replacing structure entities directly connected to similar function or benefit entities to generate new relationships and outputs novel ideas. Third, a scheme improvement process based on the co-occurrence frequency of requirement and structure knowledge supplements neglected design preferences. Final, a pipeline inspection robot case study is further employed to verify the proposed approach, and a patent knowledge graph assisted design solution prototype system is developed to assist in the utilization of innovative design knowledge. Evaluation results show the significant design potential of the proposed approach in inspiring innovative thinking and knowledge reuse.