Search

Zusammenfassung Vermehrt werden Daten in der Produktion erfasst und mit Hilfe maschineller Lernverfahren ausgewertet. Ziel dieser Datenanalysen ist es, Informationen über die Produktion zu erhalten. Typische Anwendungsbeispiele sind die Vorhersage der Produktqualität sowie die vorausschauende Wartung und Instandhaltung. Allerdings führen Datenanalysen aufgrund der domänenspezifischen Herausforderungen häufig zu keinen ausreichenden Analyseergebnissen. Die Kombination von Datenanalysen mit Expertenwissen stellt hierbei einen vielversprechenden Ansatz zur Lösung der domänenspezifischen Problemstellungen dar. Dieser Beitrag stellt hierzu eine neue Vorgehensweise für die Durchführung von Projekten des maschinellen Lernens mit wissensbasierten Ansätzen in der Produktion vor. In einem Ablaufplan werden notwendige Schritte und die Beteiligung der entsprechenden Experten vorgestellt. Für die Nutzung von vorhandenem Domänenwissen in der Produktion werden in diesem Artikel verschiedene Methoden zur Wissensmodellierung aufgezeigt und diskutiert. Der Ablaufplan und die Methoden zur Wissensmodellierung werden anschließend am Beispiel einer servopneumatischen Schweißzange validiert.

Introduction. The article describes the approach to solving the problem of complex technical system troubleshooting based on expert knowledge modeling. Intelligent information systems are widely used to solve the problems of diagnostics of multilevel systems including combine harvesters. The formal description of the subject domain knowledge is the framework for building the knowledge base of these systems. The sequence of creating an expert system knowledge base in accordance with production rules is considered. Materials and Methods. The approach is founded on the fault function table. As the object of diagnostics, one of the subsystems of the combine harvester electric equipment “opening the hopper roof flaps” is considered. The basis for constructing a sequence of elementary checks is a system of logical equations describing both the serviceable and possible faulty states of the subsystem. Results. A structural logic model is developed. As a result of analyzing the fault function table, the sets of elementary checks are determined. Four criteria have been used to analyze the weight of these checks. The authors have determined optimal sequence of checks and have developed a decision tree, which allows finding the cause of the malfunction and is the basis for creating the knowledge base of an intelligent information system. A fragment of the knowledge base is given. Discussion and Conclusion. The proposed approach of expert knowledge modelling increases the efficiency of the unit for troubleshooting of the intelligent decision support system. It makes possible to structure the base of expertise and establishing the optimal sequence of elementary checks. This allows determining the optimal sequence of application of the knowledge base production rule that makes it possible to reduce the time of restoring the serviceability of combines.

Motivation Biomedical language models produce meaningful concept representations that are useful for a variety of biomedical natural language processing (bioNLP) applications such as named entity recognition, relationship extraction and question answering. Recent research trends have shown that the contextualized language models (e.g. BioBERT, BioELMo) possess tremendous representational power and are able to achieve impressive accuracy gains. However, these models are still unable to learn high-quality representations for concepts with low context information (i.e. rare words). Infusing the complementary information from knowledge-bases (KBs) is likely to be helpful when the corpus-specific information is insufficient to learn robust representations. Moreover, as the biomedical domain contains numerous KBs, it is imperative to develop approaches that can integrate the KBs in a continual fashion. Results We propose a new representation learning approach that progressively fuses the semantic information from multiple KBs into the pretrained biomedical language models. Since most of the KBs in the biomedical domain are expressed as parent-child hierarchies, we choose to model the hierarchical KBs and propose a new knowledge modeling strategy that encodes their topological properties at a granular level. Moreover, the proposed continual learning technique efficiently updates the concepts representations to accommodate the new knowledge while preserving the memory efficiency of contextualized language models. Altogether, the proposed approach generates knowledge-powered embeddings with high fidelity and learning efficiency. Extensive experiments conducted on bioNLP tasks validate the efficacy of the proposed approach and demonstrates its capability in generating robust concept representations.

Working with cognitive concepts and propositions when encountering knowledge has become a key component in personalized skill development in higher education, which relies heavily on text (Heßdörfer, Hachmann, & Zaft, 2021; Pirnay-Dummer, 2020; Pirnay-Dummer, Haehnlein, Hachmann, & Umber, 2019). This study investigates the process of incremental text comprehension when reading new text and stepwise deepening of understanding, using two crossed factors: first, the re-representation of knowledge models derived from text as visually available artifacts to encounter in addition to text and second, tasks of stepwise conceptualization. This paradigm is designed to enable an insight into the gradual, cognitive processes characterized by increasing depth of elaboration culminating in the possible recursive reflection of text and knowledge model comprehension (possibly model revision, see Clark et al., 2012; Gentner & Stevens, 2014) at the level of single concepts and propositions (associative connections between concepts). To provide knowledge artifacts from text that have a basis in human knowledge processing (first factor), we use the knowledge modeling software T-MITOCAR (Pirnay-Dummer, 2015; Pirnay-Dummer, Ifenthaler, & Spector, 2010; Pirnay-Dummer & Lachner, 2008). This computational linguistic heuristic uses the written language’s syntax to track the associations of concepts from a text according to mental model theory (Seel, Ifenthaler, & Pirnay-Dummer, 2009; Strasser, 2010) and outputs a mathematical graph which can then be visualized as a network (Pirnay-Dummer, 2010). This factor is used to investigate whether encountering the knowledge network alone alongside text fosters text comprehension (Ballod, 2007; Mahn & Meyer, 2020). Arguably, though, for most people it should be more beneficial to be guided through the process of working with concepts and propositions (in prep.). Accordingly, we designed tasks (second factor) that directly target comprehension processes in building, shaping and re-shaping cognitive concepts and knowledge propositiones. Based on a Grounded Theory approach to interviews with experts of diverse disciplines who encountered their personal expertise text and knowledge model (in prep.), we deduced hypotheses about the exploratory and incremental comprehension process in knowledge models. They resulted in 5 stepwise concept tasks that cascade from forced-choice recognition of single concepts and single propositions through active production of concepts and propositions with a single-item contextualization until providing propositions and a short sentence to describe the knowledge context. In a 2x2 between subjects control-group design, the use of knowledge networks (network yes/no) is crossed with the stepwise concept tasks (yes/no) to predict text comprehension as measured in a comprehension task (see also Dori, Avargil, Kohen, & Saar, 2018). With the use of stepwise concept tasks, the benefit of knowledge networks should increase and interact with the benefit of concept tasks, because the networks become a visual re-representation for the task and text content at hand. In addition to immediate text comprehension questions during the same session, participants answer a delayed text comprehension task after about 2 weeks in a follow-up session. Benefits for deepended text comprehension should increase over time with consolidation, while benefits from immediate recall cease. All participants (N=80) also fulfill the MaK-adapt (reading), a test for general reading comprehension skills (Seeber, Ziegler, Frey, Balkenhol, Bernhardt, & Ebermann, 2017), the CVLT, a test for verbal learning and retention (California Verbal Learning Test; Niemann, Sturm, Thöne-Otto, & Willmes, 2008), and the CFT-20-R, a test for general fluid cognitive ability (Culture Fair Test; Weiß, 2006), as control variables. Further tasks that this sample fulfilled are the focus of investigations that go beyond this study. This study represents the main group analysis of the paradigm.

International audience; Improving surgical training by means of technology assistance is an important challenge that aims to directly impact surgical quality. Surgical training includes the acquisition of two categories of knowledge: declarative knowledge (i.e. 'knowing what') and procedural knowledge (i.e. 'knowing how'). It is essential to acquire both before performing any particular surgery. There are currently many tools for acquiring procedural knowledge, such as simulators. However, few approaches or tools allow a trainer to formalize and record surgical declarative knowledge, and a trainee to have easy access to it. In this paper, we propose an approach for structuring surgical declarative knowledge according to procedural knowledge and based on surgical process modeling. A dedicated software application has been implemented. We evaluated the concept and the software usability on two procedures with different medical populations: endoscopic third ventriculostomy involving 6 neurosurgeons and preparation of a surgical table for craniotomy involving 4 scrub nurses. The results of both studies show that surgical process models could be a well-adapted approach for structuring and visualizing surgical declarative knowledge. The software application was perceived by neurosurgeons and scrub nurses as an innovative tool for managing and presenting surgical knowledge. The preliminary results show that the feasibility of the proposed approach and the acceptability and usability of the corresponding software. Future experiments will study impact of such an approach on knowledge acquisition.

Ontology provides a shared vocabulary of a domain by formally representing the meaning of its concepts, the properties they possess, and the relations among them, which is the state-of-the-art knowledge modeling technique. However, the ontologies in the same domain could differ in conceptual modeling and granularity level, which yields the ontology heterogeneity problem. To enable data and knowledge transfer, share, and reuse between two intelligent systems, it is important to bridge the semantic gap between the ontologies through the ontology matching technique. To optimize the ontology alignment’s quality, this article proposes an Interactive Compact Genetic Algorithm (ICGA)-based ontology matching technique, which consists of an automatic ontology matching process based on a Compact Genetic Algorithm (CGA) and a collaborative user validating process based on an argumentation framework. First, CGA is used to automatically match the ontologies, and when it gets stuck in the local optima, the collaborative validation based on the multi-relationship argumentation framework is activated to help CGA jump out of the local optima. In addition, we construct a discrete optimization model to define the ontology matching problem and propose a hybrid similarity measure to calculate two concepts’ similarity value. In the experiment, we test the performance of ICGA with the Ontology Alignment Evaluation Initiative’s interactive track, and the experimental results show that ICGA can effectively determine the ontology alignments with high quality.

Emergency response is an important task in geologic hazard emergency management in China. Rapid emergency response actions can effectively reduce the losses caused by geologic hazards. Emergency plans and emergency command systems play an important role in emergency response. The emergency response still comes with challenges, such as the difficulty of integrating emergency plans with real-time geologic hazard information and the operability of emergency disposal in emergency plans. One possible approach to solving this problem is to model and represent knowledge and to enable the emergency command system to acquire and intelligently provide the required emergency response knowledge. In this paper, we propose a conceptual model of knowledge for geologic hazard emergency response (GHER). The geospatial characteristics of GHER knowledge are represented by the geo-ontology in the modeling process. The ontology enables semantic interoperability of GHER knowledge. A case study is used as an application test to evaluate the framework of GHER knowledge acquisition. In the framework, the GHER knowledge model enables the emergency command system to provide imperative knowledge intelligently for emergency decision-making; knowledge inference enables the emergency command system to provide important implicit knowledge for emergency response; and spatial analysis of geographic information system is applied to visualize and analyze spatial data in the process of emergency response.

In this paper, we present the graphical representation of the business modeling project developed for a Brazilian National Land Transportation Agency. We describe briefly an Information Asset identification, Knowledge identification and mapping, information life cycle completion and process description as part of the adopted methodology. We shows the adopted graphical representation and highlights some emerged challenges and how they have been handled. We discuss the advantages of the integrated information – knowledge graphical representation as a tool for business organization management. We argue about the need for an extension of the Unified Modeling Language(UML) towards the creation of business models that systematically Integrates knowledge and information description. Finally, we introduce new ideas on how to facilitate the building and the maintenance of the graphical representation of the models. DOI: 10.36558/rsc.v11i2.7344

Mobile augmented reality (AR) technology creates realistic learning situations and a strong sense of immersion, which is conducive to enhance learning experience and stimulate learning motivation. ...

This paper aims to develop a strategy for architectural knowledge modeling in order to actively support the built heritage conservation process by fostering collaboration among stakeholders and interoperability between datasets. The integration of two modeling systems, one ontology-based and one in BIM environment, seems to be the right way to meet this objective: the former is rather exhaustive to represent the semantic contents of conservation activities, especially non-geometrical data, the latter is absolutely suitable to represent the logic of the construction, above all geometrical-constructive aspects typical of any architectural organism. Thus, this study proposes a side-by-side approach to synchronize these different ways of representing reality by managing the complexity of cultural heritage on the one hand and of technology tools, such as information systems, on the other. The proposed methodology was tested on the city walls of San Ginesio (Macerata, Italy) and included different steps considering the in-use technologies (notably geomatics and information technologies) as key enablers to acquire, hierarchically order, model and enrich the knowledge of that heritage site. The result is a knowledge-led strategy moving from survey to HBIM implementation, as a way to enhance representation and management in architectural heritage processes.

The looming threat orbital debris poses to assets in orbit demands solutions. As the orbital population grows, so does the risk of collision, as well as the volume of associated data. It is, however, an opportunity for interdisciplinary innovation and cooperation. This paper focuses on the data management and knowledge modeling aspect of developing solutions for a sustainable and safe orbital space environment. An in-progress work to develop an orbital debris reference ontology [Rovetto, 2015] is summarized in order to discuss knowledge modeling for orbital debris. The formal analysis of this effort can contribute to standards development, as well as terminological and policy challenges. Leveraging the growing volumes of orbital debris and space situational awareness (SSA) data will create a more complete picture of the orbital space environment. Part of the solution will be: consistent and correct data interpretation, sharing orbital debris and SSA data in one form or another, terminology development & harmonization, and knowledge or domain modeling. To facilitate this, [Rovetto, 2015] proposed ontology development for the orbital debris and broader space domain. This paper summarizes concepts from that paper, and subsequently developed concepts [2-9]. See also https://purl.org/space-ontology and https://purl.org/space-ontology/odo . Ontology engineering is an interdisciplinary area of research related to knowledge representation and reasoning in artificial intelligence, model-based systems engineering, semantic technologies and the so-called semantic web. An ontology is effectively a computable and semantically rich terminology that presents a knowledge or domain model for a topic area. Expressions of knowledge, beliefs, or assertions are stored using formally defined terms. This knowledge base is reasoned over to yield answers to database queries, among other things. Ontologies have been developed in knowledge-based projects across various disciplines, and used for such things as search engines, chatbots, and enterprise knowledge graphs. Ontologies aim to support: interoperability, automated reasoning, data sharing and integration, data search and retrieval, and communicating the meaning of data. The Orbital Debris Ontology (ODO) [1], and related ontologies [Rovetto & Kelso 2016] [Rovetto 2016, 2017], were proposed to help achieve this. ODO, for instance, is intended as a domain ontology that can be used across federated databases, offering an explicitly specified set of concepts describing the orbital debris domain. Its meaning-rich taxonomy will provide a sharable semantics for orbital debris data to, in part, consistently communicate the meaning of data to both humans and machines, and tag data elements in space object catalogs to help afford automated reasoning tasks, decision support, knowledge discovery, and information integration. ODO and the SSA ontology (SSAO) [2] is part of the overall Orbital Space Ontology concept, which is conceived as a unifying domain reference ontology. It aims to provide a knowledge representation structure of orbital space, a common semantic model, and develop a sharable terminology. Collectively this will provide common meaning for datasets, a high-level taxonomy or classification for orbital space objects, and a means to characterize space objects. Ongoing efforts have included using visualizations, R, JSON-LD, and contemporary semantic technologies. Potential applications include web-based platforms, web apps, visualizations, modeling and simulations. Community input and participation may yield a more widely understood domain model as well as facilitate terminological standards. For example, conceptual, terminological, and ontological analysis can make helpful contributions to such efforts as the Space Debris Mitigation Requirements by developing more precise, consistent and coherent terms and definitions. Other projects can also use ODO (and its related ontologies) as a common knowledge model, metadata set, taxonomy or vocabulary. Readers interested in supporting development are encouraged to make contact.

PurposeHealth-care ontologies and their terminologies play a vital role in knowledge representation and data integration for health information. In health-care systems, Internet of Technology (IoT) technologies provide data exchange among various entities and ontologies offer a formal description to present the knowledge of health-care domains. These ontologies are advised to assure the quality of their adoption and applicability in the real world.Design/methodology/approachOntology assessment is an integral part of ontology construction and maintenance. It is always performed to identify inconsistencies and modeling errors by the experts during the ontology development. A smart health-care ontology (SHCO) has been designed to deal with health-care information and IoT devices. In this paper, an integrated approach has been proposed to assess the SHCO on different assessment tools such as Themis, Test-Driven Development (TDD)onto, Protégé and OOPs! Several test cases are framed to assess the ontology on these tools, in this research, Themis and TDDonto tools provide the verification for the test cases while Protégé and OOPs! provides validation of modeled knowledge in the ontology.FindingsAs of the best knowledge, no other study has been presented earlier to conduct the integrated assessment on different tools. All test cases are successfully analyzed on these tools and results are drawn and compared with other ontologies.Originality/valueThe developed ontology is analyzed on different verification and validation tools to assure the quality of ontologies.

Intelligent education systems have enabled personalized learning (PL). In PL, students are presented with educational contents that are consistent with their personal knowledge states (KS), and the critical task is accurately estimating these states through data. Knowledge tracing (KT) infers KS (latent) through historical student interactions (observed) with the knowledge components (KCs). A wide variety of KT techniques have been developed, from Bayesian Knowledge Tracing (BKT) to Deep Knowledge Tracing (DKT). However, in most of these methods, the KCs are represented as stand-alone entities, and the effect of representing KCs using contexts such as learning-related factors has been under-investigated. Also, KT needs to generate personalized results to facilitate tasks such as exercise recommendation. In this paper, we propose two approaches that use a contextualized representation of KCs, one with a content-based approach and another with a Long Short Term Memory (LSTM) network plus a personalization mechanism. By performing extensive experiments on two real-world datasets, results show not only a tangible improvement in prediction accuracy in the KT task compared to existing methods, but also its effectiveness in improving the recommendation precision.

Assistive health care system is a viable solution for elderly care to offer independent living. Such health care systems are feasible through smart homes, which are intended to enhance the living quality of the occupant. Activities of daily living (ADL) are considered in the design of a smart home and are extended to abnormality detection in the case of health care. Abnormality in occupant behavior is the deviation of ongoing activity with that of the built activity model. Generally, supervised machine learning strategies or knowledge engineering strategies are employed in the process of activity modeling. Supervised machine learning approaches incur overheads in annotating the dataset, while the knowledge modeling approaches incur overhead by being dependent on the domain expert for occupant specific knowledge. The proposed approach on the other hand, employs an unsupervised machine learning strategy to readily extract knowledge from unlabelled data using contextual pattern clustering and subsequently represents it as ontology activity model. Ontology offers enhanced activity recognition through its semantically clear representation and reasoning, it has restriction in handling temporal data. Hence, this article in addition to unsupervised modeling focuses at enabling temporal reasoning within ontology using fuzzy logic. The proposed fuzzy ontology activity recognition (FOAR) framework represents an activity model as a fuzzy temporal ontology. Fuzzy SWRL rules modeled within ontology aid activity recognition and abnormality detection for health care. The experimental results show that the proposed FOAR has better performance in abnormality detection than that of the existing systems.

The data are evolved from heterogeneous sources as structured, unstructured, and semi-structured in varied formats from various platforms and applications, extending data integration difficulty. Data heterogeneity in varied formats and representation leads to a standardization gap that has to be addressed on enabling common data representation. Semantic web technologies are used to address the gap of data interoperability by representing data in a machine-understandable format for knowledge inference defined as taxonomy, thesaurus, and ontology to infer knowledge. An end-to-end knowledge graph is proposed to integrate diverse data generated from heterogeneous data sources in this work. A brief view of the data integration process and NLP techniques are discussed to identify concepts from unstructured data inputs to the knowledge representation ontology. The concepts for knowledge modeling are described as a logistic domain for shipping processes. Logistic information is extracted for conceptualization and represented as taxonomy. Ontology is modeled for managing logistic shipping containers to infer knowledge of containers used in the logistic shipping process. The proposed knowledge base is evaluated with a specific use case to infer knowledge for cargo tracking in shipment processes.

Ontology technology has been investigated in a wide range of areas and is currently being utilized in many fields. In the e-learning context, many studies have used ontology to address problems such as the interoperability in learning objects, modeling and enriching learning resources, and personalizing educational content recommendations. We systematically reviewed research on ontology for e-learning from 2008 to 2020. The review was guided by 3 research questions: “How is ontology used for knowledge modeling in the context of e-learning?”, “What are the design principles, building methods, scale, level of semantic richness, and evaluation of current educational ontologies?”, and “What are the various ontology-based applications for e-learning?” We classified current educational ontologies into 6 types and analyzed them by 5 measures: design methodology, building routine, scale of ontology, level of semantic richness, and ontology evaluation. Furthermore, we reviewed 4 types of ontology-based e-learning applications and systems. The observations obtained from this survey can benefit researchers in this area and help to guide future research.

The collaborative approach which are born OER (Open Educational Resource) is the joint search for pedagogical and technological challenges to achieve improved quality reconstruction. It is necessary to demonstrate defined metrics that support processes and provide guidelines or instructions related to design for all. The open feature of the OERs must demonstrate accessibility, guaranteeing their interoperability. The incorporation of intelligent systems based on the efficient publication of accessibility information, contributes to the evaluation of accessible resources, knowledge modeling and future trends based on the feedback of profiles, and personalization from the user experience through. This paper presents the proposal for an expert system based on the ISO 40500, ISO 24751 and UDL standards.

The paper presents an attempt of real, industrial, engineering knowledge modeling, used in a small company. The authors concentrate on the concept of personal/team knowledge repositories and tools which can support conceptual design. The performance of engineering design tasks usually concentrates on two types of activities: activities connected with modeling the reality, activities analysing the models of reality and associated decision-making actions. Both types of activities are normally fulfilled and integrated by the designers. In doing so, the applied models can be at different levels of complexity, starting from very simple ones, which are based on a few formulae to huge systems of equations. Problems of a high degree of complexity are mostly solved in collaboration with other designers. All activities mentioned above require engineering knowledge which is the result of education and professional experience. The designer is able to operate actively with his knowledge. The knowledge can be the result of reflection on past tasks which he recalls. In many cases, especially in conceptual design, this is a kind of evolutionary progress in a selected knowledge development. In the paper the authors concentrate on the conceptual design, its functioning, its personal/team knowledge development/evolution contexts, its knowledge background and knowledge roots. The whole approach was created and verified in a real industrial design office. The introduced software solutions allow to perform, acquire and support key steps which are necessary to achieve the final conceptual goals.

In shipbuilding, forming plate by line heating is atypical complex manufacturing process involving many uncertain factors, so it is difficult to establish an accurate mathematical model. How to establish a knowledge model that can reflect technological laws is the key to the development of an intelligent decision system for forming plate by line heating. In this article, rough set (RS) theory is applied to the modeling of the line heating process. By defining variable-inclusion RSs, an algorithm of knowledge reduction is proposed, which enhances the noise immunity and fault tolerance of the model, and improves the efficiency of knowledge acquisition. Through introducing fuzzy logic, a method of modeling the line heating process based on RSs and fuzzy logic is proposed, which effectively extracts the technological rules of plate formation. Finally, rapid decision-making for process parameters is implemented by fuzzy inference technology. 1. Introduction For some complex manufacturing technologies in modern shipbuilding, such as line heating and welding processes, it is difficult to establish an exact mathematical model because of high nonlinearity, multivariable coupling, and uncertainties of the system. With the development of computer technology and artificial intelligence, soft computing methods such as artificial neural networks, genetic algorithms, fuzzy logic, and rough set (RS) theory have been applied successively in ship manufacturing process modeling, which shows good prospects for intelligent technology in shipbuilding. Shin et al. (1999) at Seoul National University used a single-curvature plate model to simulate the formation of saddle-type shells and deduce the technological parameters of line heating by an artificial neural network. They also proposed a comprehensive algorithm for automatically curving plate by line heating, and further developed an application system that can simulate the deformation of double-curved plates (Shin et al. 2004a, 2004b). Liu et al. (2006) applied a hierarchical genetic algorithm to optimize the technological parameters of an automatic line heating process. In the field of ship welding, fuzzy logic technology has been used to establish a fuzzy model of the relationship between welding variables and weld forming-parameters (Su 2009). Feng (2012) set up a knowledge base of a ship-welding process by a RS method and then implemented ship-welding production design through uncertainty reasoning. Based on RS theory, Chen and Lv (2013) developed a data-driven knowledge base for quality control of ship hull welding.

The use of Knowledge Engineering (KE) processes to analyze and configure domains in automated planning is becoming more appealing since it was noticed that this issue could make a difference to solve real problems. The contrast between a generic domain independent approach, taken as canonical in AI, and alternative processes that include knowledge engineering – eventually adding specific knowledge – has been discussed by Computer and Engineering communities. A big impact has been noticed mainly in the early phase of requirement analysis when KE approach is normally introduced. Requirement analysis is responsible for carrying out the Knowledge modeling of both problem and work domains, which is a key issue to guide different planner algorithms to come out with efficient solutions. Also, there is the scalability issue that appear in most real problems. To face that, hierarchical methods played an important hole in the history of planning and inspired several solutions since the proposal of NONLIN in the 70’s. Since then, the idea of associating hierarchical relational nets with partial ordered actions has prevailed when large systems were considered. However, there is still a gap between the hierarchical approach and the state of art of requirements analysis to allow features anticipated by KE approach to really appear in the requirements of a planning process. This paper proposes a pathway to solve this gap starting with requirements elicitation represented first in the conventional semi-formal (diagrammatic) language – UML – that is translated to Hierarchical Petri Nets (HPNs) by a new enhanced algorithm. The proposed process was installed in a software tool – developed by one of the authors – that analyzes the performance of the KE planning model: itSIMPLE (Integrated Tools Software Interface for Modeling Planning Environment). This tool was initially designed to use classic Place/Transition nets and an old version of UML (2.1). It is now enhanced to use UML 2.4 and a hierarchical Petri Net extension, also developed by the authors. Realistic examples illustrate the process which is now being applied to larger problems related to the manufacturing of car sequencing domain, one of challenge of ROADEF 2005 (French Operations Research & Decision Support Society). Finally, we consider the possibility to introduce another approach to the KE process by using KAOS (Keep All Object Satisfied) to make the planning design more accurate.

This article presents a dynamic hazard identification methodology founded on an ontology-based knowledge modeling framework coupled with probabilistic assessment. The objective is to develop an efficient and effective knowledge-based tool for process industries to screen hazards and conduct rapid risk estimation. The proposed generic model can translate an undesired process event (state of the process) into a graphical model, demonstrating potential pathways to the process event, linking causation to the transition of states. The Semantic web-based Web Ontology Language (OWL) is used to capture knowledge about unwanted process events. The resulting knowledge model is then transformed into Probabilistic-OWL (PR-OWL) based Multi-Entity Bayesian Network (MEBN). Upon queries, the MEBNs produce Situation Specific Bayesian Networks (SSBN) to identify hazards and their pathways along with probabilities. Two open-source software programs, Protege and UnBBayes, are used. The developed model is validated against 45 industrial accidental events extracted from the U.S. Chemical Safety Board's (CSB) database. The model is further extended to conduct causality analysis.

Today's educational systems need an efficient tool to perform competently assessment of students on their major concepts they learnt from study material. Preparing a set of questions for assessment can be time consuming for teachers. Automatic Question Generation (AQG) is the technique for generating a right set of questions from a content. A key problem with AQG systems is the selection of the appropriate target sentences and concepts. In this paper, we investigate the difficulties in automatic target selection using unannotated document sources. As the result of the analysis, we identified the following key issues: hard to measure the topic relevance; the large segmentation of the information content and the incomplete background knowledge bases.

Integrating external knowledge into the recommendation system has attracted increasing attention in both industry and academic communities. Recent methods mostly take the power of neural network for effective knowledge representation to improve the recommendation performance. However, the heavy deep architectures in existing models are usually incorporated in an embedded manner, which may greatly increase the model complexity and lower the runtime efficiency. To simultaneously take the power of deep learning for external knowledge modeling as well as maintaining the model efficiency at test time, we reformulate the problem of recommendation with external knowledge into a generalized distillation framework . The general idea is to free the complex deep architecture into a separate model, which is only used in the training phrase, while abandoned at test time. In particular, in the training phrase, the external knowledge is processed by a comprehensive teacher model to produce valuable information to teach a simple and efficient student model. Once the framework is learned, the teacher model is abandoned, and only the succinct yet enhanced student model is used to make fast predictions at test time. In this article, we specify the external knowledge as user review, and to leverage it in an effective manner, we further extend the traditional generalized distillation framework by designing a Selective Distillation Network (SDNet) with adversarial adaption and orthogonality constraint strategies to make it more robust to noise information. Extensive experiments verify that our model can not only improve the performance of rating prediction, but also can significantly reduce time consumption when making predictions as compared with several state-of-the-art methods.

Knowledge plays a very important role in the dialogue systems. Inspired by how humans use unstructured background knowledge in the conversations, this paper proposes a dialogue generation model based on multi-task learning. The model divides the conversation generation task into two tasks, a knowledge selection task and a response prediction task, which are regard as a reading comprehension task and a text generation task separately. Specifically, in the task of knowledge selection, a language pre-training model Bidirectional Encoder Representations from Transformers (BERT) is applied to solve the problem of selecting the knowledge from the background knowledge documents in the current context. And in the task of response prediction, a transformer version of pointer-generator network, being composed of an encoder using the shared BERT mentioned in knowledge selection and a decoder using the left-context-only transformer, is applied to copy tokens from the background knowledge via pointing and produce tokens in the vocabulary through a generator. Our experiments on the HOLL-E dataset show that our model achieves better results than the strong baseline models and the related recent work.

Knowledge tracing (KT) is a research topic which seeks to model the knowledge acquisition process of students by analyzing their past performance in answering questions, based on which their performance in answering future questions is predicted. However, existing KT models only consider whether a student answers a question correctly when the answer is submitted but not the in-question activities. We argue that the interaction involved in the in-question activities can at least partially reveal the thinking process of the student, and hopefully even the competence of acquiring or understanding each piece of the knowledge required for the question. Based on real student interaction clickstream data collected from an online learning platform on which students solve mathematics problems, we conduct clustering analysis for each question to show that clickstreams can reflect different student behaviors. We then propose the first clickstream-based KT model, dubbed clickstream knowledge tracing (CKT), which augments a basic KT model by modeling the clickstream activities of students when answering questions. We apply different variants of CKT and compare them with the baseline KT model which does not use clickstream data. Despite the limited number of questions with clickstream data and its noisy nature which may compromise the data quality, we show that incorporating clickstream data leads to performance improvement. Through this pilot study, we hope to open a new direction in KT research to analyze finer-grained interaction data of students on online learning platforms.

Learning from graph and relational data plays a major role in many applications including social network analysis, marketing, e-commerce, information retrieval, knowledge modeling, medical and biological sciences, engineering, and others. Recently, Graph Neural Networks (GNNs) have emerged as a promising new learning framework capable of bringing the power of deep representation learning to graph and relational data. This ever-growing body of research has shown that GNNs achieve state-of-the-art performance for problems such as link prediction, fraud detection, target-ligand binding activity prediction, knowledge-graph completion, and product recommendations. In practice, many of the real-world graphs are very large. It is urgent to have scalable solutions to train GNN on large graphs efficiently. The objective of this tutorial is twofold. First, it will provide an overview of the theory behind GNNs, discuss the types of problems that GNNs are well suited for, and introduce some of the most widely used GNN model architectures and problems/applications that are designed to solve. Second, it will introduce the Deep Graph Library (DGL), a scalable GNN framework that simplifies the development of efficient GNN-based training and inference programs at a large scale. To make things concrete, the tutorial will cover state-of-the-art training methods to scale GNN to large graphs and provide hands-on sessions to show how to use DGL to perform scalable training in different settings (multi-GPU training and distributed training). This hands-on part will start with basic graph applications (e.g., node classification and link prediction) to set up the context and move on to train GNNs on large graphs. It will provide tutorials to demonstrate how to apply the techniques in DGL to train GNNs for real-world applications.

Multi-variable and non-linear features for wastewater treatment process (WTP), traditional expert experience-based decision making scheme can not well express the WTP. To overcome the challenge, this paper proposes to build Robotic Decision scheme based on joint awareness of knowledge and data for modeling of WTP. Firstly, Modeling WTP by knowledge modeling system and neural network modeling system respectively. After that, Establish a collaboration awareness layer to ingeniously combine knowledge model with data model. At last, a series of experiments are carried out on a dataset obtained from a realistic wastewater treatment plant, in order to evaluate the performance and reliability of the proposed RD-ACL. The results show that the proposal improves modeling efficiency compared with typical existing methods. Through such a collaborative modeling scheme, intelligent management degree of WTP can be improved to some extent.

The design and construction of personalized and flexible e-learning systems can benefit from the use of sophisticated knowledge modeling and data-analysis techniques to provide learning resources that are accessible by anyone, including learners with special accessibility needs and preferences. The paper describes a personalized educational resources system that considers learner's needs, especially when it comes to learners with accessibility requirements. We use Semantic Web technologies as resource representation formalism to provide learners with suitable learning objects, learning activities, and teaching methods based on their different preferences and needs. The system supports the creation of interactive semantic queries to bring the result sets closer to user research requirements.

Building upon the clinical evidence supporting that decomposing a pathological image into different components can improve diagnostic value, in this paper we propose a Decomposition-and-Fusion Network (DFNet) for HE-stained pathological image classification. The medical goal of using HE-stained pathological images is to distinguish between nucleus, cytoplasm and extracellular matrix, thereby displaying the overall layouts of cells and tissues. We embed this most basic medical knowledge into a deep learning framework that decomposes a pathological image into cell nuclei and the remaining structures (that is, cytoplasm and extracellular matrix). With such decomposed pathological images, DFNet first extracts independent features using three independent CNN branches, and then gradually merges these features together for final classification. In this way, DFNet is able to learn more representative features with respect to different structures and hence improve the classification performance. Experimental results on two different datasets with various cancer types show that the DFNet achieves competitive performance.

Knowledge modeling is required to analyze the properties and relationships in a specific domain. Several mechanisms are used to represent knowledge; one of them is the logical rule that is an intuitive way to represent schemes through a structure that facilitates matching a premise with a corresponding conclusion.

The article introduces the problem of maximum simplification of the process of information visualization and automation of design and production processes using "big data," which helps to analyze trends based on analytics and forecast stocks, production volumes, service life, and cycles of equipment operation, etc. This article discusses the concept of prediction models and their application. The article also provides an introduction to the concept of proactive error prevention in production. The challenges for research include prediction accuracy, the automated adaptation of prediction models to a variety of processes, and the management of prediction models in changing processes. The proposed solution is a combination of machine learning technologies and Big Data-based approaches, as well as knowledge modeling.

The ontology has marked its presence in several research fields in order to address issues in knowledge modeling’ Artificial Intelligence, classification, clustering and more specifically in knowledge engineering. In the Web Service field, ontology is used principally to describe the Web Service interface for discovering, storing and composing Web services. This is well known as semantic Web services, such as OWL-S, WSMO and WSDL-S. Unfortunately, work in this field remain insufficient till now; so, supporting this description by adding a shared ontology (knowledge base) can improve the performance of the Web Service technology. Manual construction of ontology is a tedious and expensive task which requires domain expert intervention; an automatic ontological construction process will be interesting and less costly solution in terms of time and money. In this paper, we propose to use clustering algorithms to create an ontological network (Service Ontology (SO)) from a semantic Web Services corpus. The created SO can be used for storage and discovery of Web Services in a distributed and intelligent environment.

Realization of a cloud enabled Knowledge modeling as a Service (KaaS) supports acquiring, managing and sharing knowledge through different types of sources. Cloud computing enables its users to access and edit their data easily without any administration and software development expenses and difficulties. In this paper, an overview and implementation results of the concept called Knowledge Forge is described. The proposed concept of this research is successfully tested and applied in the representation and management of job knowledge (competences) towards the assessment of talent's proficiency. This can also be used for modeling of graduates and students proficiency in order to identify their competence gaps and assign proper learning sources to them in order to improve their competitiveness. The Knowledge Forge can be extended easily to a wide variety of applications in order to enable further knowledge formats and application areas. Thereby, new formats use their own graphical interface and methods, while no changes in the system itself are required. The main goal of this research is to enable skilled workers and knowledge engineers to search for, modify and connect different formats of knowledge(e.g. unstructured knowledge). Neither users nor administrators have to care about the data format, so a search may show better results out of all stored data along with external search results provided by web search engines. The result of this work is a framework for all KaaS purposes. Although its use is limited at the moment it creates a basis for new applications in several areas. Further work can enhance the functionality or simplify the enhancement process to enable even the users to create their own applications.

Fuzzy logic is an extension of Boolean logic created by Lotfi Zadeh in 1965 based on his mathematical theory of fuzzy sets, which is a generalization of classical set theory. By introducing the notion of degree or possibility, fuzzy logic confers very appreciable flexibility to the reasoning, which uses it, which makes it possible to take into account imprecision and uncertainties. Despite a beginning of reluctance and even rejection of the theory of fuzzy logic, it has been used in many fields. Today, its usefulness and its reputation are no longer to be demonstrated because more than 50 years after its appearance, it has been well adopted by engineers and part of the scientific world. In this paper, we present a state of the art on fuzzy logic and its use in the field of knowledge representation and more specifically in ontology modeling using OWL and SWRL.

The ornamental art found in Moroccan architecture is revealing ingenuity and a very valuable know-how that has been passed from generation to generation in the secrecy of the workshops and in the absence of documentation. The technique widely used for geometric drawing of patterns is called “Hasba” and its description remains ambiguous hence the interest to formalize the method construction rules through a formal modeling of the knowledge involved. The method used for this purpose is MASK, which has been chosen for its many advantages (ease of appropriation, flexibility, generic aspect, etc.). Its application to the field of Moroccan craftsmanship is the subject of this paper. It has been realized through a case study with expert craftsmen. However, this experiment revealed several challenges that are largely due to the informal nature of the craftsmanship domain and the artisans’ profile. Thus, an adaptation of the methodological framework was necessary and allowed us to validate our approach. The result is a set of interleaved and interconnected models constituting a small “knowledge book” on the geometric drawing technique called “Hasba”.

Finding relevant expertise is a critical need in software organizations since developers use it to support their knowledge needs. In software organizations, expertise can be found in architectural knowledge (AK), which incorporates the experience and problem reasoning of human resources through two primary sources: experts and artifacts; consequently, know how to model it, could grant access to the best practices, training practices, and data across applications; this has generated considerable interest by organizations, which seek to develop formal approaches to condensate the available knowledge from different sources in a systematic manner. In this work, we present a semantic knowledge modeling, which was developed using knowledge modeling techniques; the model aims to link artifacts and experts within the organization; moreover, we present ExCap a tool that implements the semantic model presented.

As a hot research field of big data intelligence, knowledge graph is widely concerned and discussed in recent years. This thesis elaborated the definition and architecture of knowledge graph, the development drive and its importance, analyzed the seven technologies in the knowledge graph: knowledge acquisition, knowledge representation, knowledge storage, knowledge fusion, knowledge modeling, knowledge computation and knowledge operation and maintenance, and introduced and prospected the application of knowledge graph in the agricultural information consultation and the challenges faced in its future development.

Textbooks are educational documents created, structured and formatted by domain experts with the main purpose to explain the knowledge in the domain to a novice. Authors use their understanding of the domain when structuring and formatting the content of a textbook to facilitate this explanation. As a result, the formatting and structural elements of textbooks carry the elements of domain knowledge implicitly encoded by their authors. Our paper presents an extendable approach towards automated extraction of this knowledge from textbooks taking into account their formatting rules and internal structure. We focus on PDF as the most common textbook representation format; however, the overall method is applicable to other formats as well. The evaluation experiments examine the accuracy of the approach, as well as the pragmatic quality of the obtained knowledge models using one of their possible applications -- semantic linking of textbooks in the same domain. The results indicate high accuracy of model construction on symbolic, syntactic and structural levels across textbooks and domains, and demonstrate the added value of the extracted models on the semantic level.

In many games, NPC-player interactions play a vital role in gameplay. Previous literature has successfully shown how NPC interaction focused on social simulation is an effective means for creating dynamic characters such as in the games Prom Week and Versu. We believe that social simulation is a key element in the creation of complex characters, which is further aided by natural language interaction and knowledge modeling. In this work, we propose an architecture for player-NPC interactions built on top of the Ensemble engine that additionally incorporates chatbots and knowledge modeling technology, with the objective of making craftable and interesting NPCs more easily authorable.

Since the power equipment related data is distributed in different information systems of the power system, the required information to construct a complete power equipment supply chain is complex and disperse. The description of transportation and management in the equipment supply chain can be defined as multi-dimensional data. A series of entities with multidimensional data can establish a hierarchy to describe the affiliation relationship in management and geographic location in a supply chain. In the hierarchy ontology, each hierarchy forms a sequence of node data for internal information analysis and the cross-layer data can be queried through relationship links. In this study, a structural modeling method of equipment supply chain based on the ontology hierarchy framework is proposed. The multi-dimensional feature in the whole process of power equipment production, sales, transportation, and management can be used to establish and record an organized and sustainable transmission process. By comparing with the general intuitive ontology model in terms of construction complexity and query performance, the reliability and efficiency of the hierarchical structure are determined.

Learning from graph and relational data plays a major role in many applications including social network analysis, marketing, e-commerce, information retrieval, knowledge modeling, medical and biological sciences, engineering, and others. In the last few years, Graph Neural Networks (GNNs) have emerged as a promising new supervised learning framework capable of bringing the power of deep representation learning to graph and relational data. This ever-growing body of research has shown that GNNs achieve state-of-the-art performance for problems such as link prediction, fraud detection, target-ligand binding activity prediction, knowledge-graph completion, and product recommendations. In practice, many of the real-world graphs are very large. It is urgent to have scalable solutions to train GNN on large graphs efficiently. The objective of this tutorial is twofold. First, it will provide an overview of the theory behind GNNs, discuss the types of problems that GNNs are well suited for, and introduce some of the most widely used GNN model architectures and problems/applications that are designed to solve. Second, it will introduce the Deep Graph Library (DGL), a scalable GNN framework that simplifies the development of efficient GNN-based training and inference programs at a large scale. To make things concrete, the tutorial will cover state-of-the-art training methods to scale GNN to large graphs and provide hands-on sessions to show how to use DGL to perform scalable training in different settings (multi-GPU training and distributed training). This hands-on part will start with basic graph applications (e.g., node classification and link prediction) to set up the context and move on to train GNNs on large graphs. It will provide tutorials to demonstrate how to apply the techniques in DGL to train GNNs for real-world applications.

Knowledge modeling, representation, and automatic knowledge inference are fundamental problems for developing automatic situation awareness (ASA) systems in the field of complex power system rescheduling. In this paper, we present a knowledge representation model named dynamic rescheduling graph (DRG), i.e., a new kind of relational knowledge graph, to represent the knowledge of the power dispatchers. DRG has many advantages, such as using different knowledge representation methods like function and logic expression and setting attributes to nodes and edges, which make it easier to develop an ASA system for complex situation diagnose, event solving, and rescheduling. To make inference timely in the DRG, we develop a unique algorithm called hybrid mapping inference (HMI), which supports forward, backward, and mapping inference. We also build an initial ASA system based on the proposed knowledge model and the inference algorithm. We validate the effectiveness of the ASA system in the Guangzhou regional power grid.

The application of the knowledge graph in computer vision is a new trend in deep learning. Facial video-based emotion analysis and recognition are critical topics of research in the mental healthcare field. In this paper, we proposed a novel noncontact intelligent framework to represent the knowledge of facial features and heart rate (HR) features for predicting the emotional states of objects. The framework is divided into two parts: knowledge modeling and knowledge reasoning. In the first step of knowledge modeling, 3D-CNN is utilized to model the spatiotemporal information from the facial and forehead regions based on the remote photoplethysmography technique, separating the blood volume pulse (BVP) signal and extracting the HR from the forehead image sequence. Finally, the multichannel features are integrated and transformed into structured data and put into the knowledge graph as much as possible. Knowledge reasoning is an inferential process that associates the deep learning model with structured knowledge to predict continuous values of the emotional dimensions (pleasure, arousal, and dominance) from facial videos of subjects. Experiments conducted on the DEAP database demonstrate that this approach leads to improved emotion recognition performance and significantly outperforms recent state-of-the-art proposals. The result proved that prior knowledge from the knowledge graph ground truth on deep learning is an efficient means of emotional recognition in vision modality. Our artificial intelligence models can be popularized and applied in daily healthcare.

In order to do a good job in providing information services for emergency response to emergencies, efficiently handle and respond to various emergencies. This paper presents a method of knowledge modeling for emergency management in information service domain. The ontology of emergency information service field is researched. The hierarchical and non-hierarchical relationships between classes which in the ontology of information service field is analyzed. The construction of the ontology model of the emergency disposal information service for blocking the river bank breach is investigated. The processe of constructing the ontology model of the emergency response is presented in detail. The experimental results show that an ontology model of emergency response in the information services field is able to describe the relationship between information service activities, information resources, and combat forces in the emergency response process. Besides, it can obtain the inner information of the relationships between service activities through reasoning and make predictions about the requirements for information resource in emergency managements. This method has potential to provide auxiliary decision support for the intelligent development of information service work.

In 2008, FPT corporation launched the Internet-Violympic Math Competition, the first contest in Vietnam that combines modern educational methods and strengths in IT, helping high school students love and learn more effectively. With nearly 30 million contestants in 63 provinces and cities across the country, the competition has created the joy of learning for tens of millions of students over the past 10 years. Following the success of Violympic, FPT corporation brought the VioEdu Online Education System to the market, contributing to creating fair learning opportunities for students nationwide, promoting the quality of national education. VioEdu is an online training system for elementary, junior high and high school students, helping to personalize their learning, save time and increase learning efficiency by applying the latest technology such as: Artificial intelligence (AI), Big data analysis, graphical knowledge modeling and compatible learning methods. Accordingly, the training process, data on learning behaviors and knowledge of students will be synthesized and analyzed by AI to find strengths and gaps in knowledge to propose appropriate learning pathways. As a result, each skill, depending on the nature, will be cultivated or promoted separately, helping them to learn the right focus and not spread out. The content of the lessons on VioEdu is developed and developed by a team of excellent teachers and educational experts from education departments and universities, following the curriculum of Vietnamese textbooks. In addition, the system also has about 30% of advanced and expanded knowledge. The form of visual lecture presentation, built as a game with encouraging prizes, creates a learning interest for children. Only with computers or mobile devices connected to the internet, students from all over the country can access modern education, shorten the knowledge gap between regions by the VioEdu online education system. Currently, the system has more than 400 video lectures and more than 100,000 content of Math teaching knowledge for elementary students. In the product development roadmap, VioEdu will cover all content of all levels of Vietnamese Mathematics, then extend to Mathematics in English, English, Physics, Chemistry, and STEM. The paper attempts to evaluate the effectiveness of the VioEdu system based on the testing process in the 2018-2019 school year with 2,000 students at some elementary schools in Hanoi, with 73.8% of students making progress. after 3-6 months of studying on VioEdu (according to an independent objective assessment of the teacher directly teaching).

Since Google proposed the concept of the knowledge graph in 2012, the research interest of the knowledge graph has been increased year by year. From the construction of more and more general knowledge graph to the hot research of domain knowledge graph in recent years, it will found that people have more in-depth research on knowledge, more refined granularity, and more stringent requirements on quality. In teaching, it is also essential to construct a course knowledge graph to assist in practical education. This article takes the new curriculum standard for the high school information technology compulsory book one "Data and Computing" and compulsory book two "Information System and Society" as examples. Based on the textbook organization, it designed a course knowledge graph that includes three types of knowledge units: chapters, sections, and knowledge points, and three semantic relationships: sequences, contains, and connects. Through the analysis of the basic structure and specific construction process of the course knowledge graph, this paper explores that the course knowledge graph can well support systematic learning, personalized learning, and other learning situations.