Search

We create a Knowledge Graph for Humanities research. Starting with a multidisciplinary dataset of 25,000 OCRed JSTOR papers, we use Deep Learning methods to filter out OCR noise, extract and interrelate research activities, methods and goals, associate them with metadata and transform each paper into approximately 200 RDF triples.

We formulate a domain-independent language for representing stochastic tasks, and show how this representation can be synthesized from few training observations. In one experiment, we study how a physical robot may learn to fold clothes from a small number of visual observations, in contrast to big data techniques. Under the same framework, we also show how a virtual chat-bot may learn dialogue policies from few example transcripts, resulting in an interpretable dialogue model, outperforming current statistical techniques. Central to both examples in the concept of utility, why it's essential for generalizability, and how to learn it from small data.

Introducing Artificial Intelligence (AI) tools is one of the development trends in complex industrial systems in the industry 4.0 environment. Unique challenges in system operations need to be handled by effective operation support systems. The knowledge-based operation support systems are developing rapidly in recent years. The paper aims at highlighting the concerns of knowledge acquisition and representation in one of the knowledge-based methodologies, the Multilevel Flow Modelling (MFM). A procedure of knowledge acquisition and representation for building MFM models is proposed to aim at improving the overall model quality and consistency. An interface linking systems' instrumentations to MFM functions are introduced. The new reasoning engine is used for MFM based real-time cause-consequence reasoning about dynamic plant situations. The model verification and validation, and the model performance evaluation analysis method are proposed. This paper also provides case studies that illustrate the effectiveness of intelligent operation support by applying MFM to an off-shore water injection system. It demonstrates that the procedure of knowledge acquisition and representation can facilitate the model builders, and ensure the quality of the models used for operation support. (c) 2021 The Author(s). Published by Elsevier B.V. on behalf of Institution of Chemical Engineers.

There is a crucial need for tools to help designing sustainable agroecosystems based on agroecology. Indeed, such agroecosystems are inherently complex and their design requires to integrate various data and unstabilized scientific knowledge. In this thesis, we consider the issue of selecting service plant species according to their potential to provide ecosystem services.To tackle that issue, we adopt an approach based both on a formalized representation of domain knowledge, which enables reasoning, and on the exploitation of available data, collected independently of the targeted application. More specifically, we rely on the one hand on recent scientific results in agronomy linking functional traits (i.e., measurable charactericsof plant species) to ecosystem services, and on the other hand on data about functional traits collected by the research community in ecology. The architecture of our system is inspired by the ontology-based data access paradigm, which allows to combine data and knowledge in a principled way. We provide a methodology to acquire scientific knowledge in the form ofdiagrams linked to data sources, as well as a formalization in a logical rule-based language. Importantly, our rules are independent from specific diagrams and data, to ensure genericity and facilitate the evolution of the system. We implement our methodology in the case study of vine grassing, i.e., installing herbaceous service plants in vineyards. We provide an evaluation of the system’s results on this use case and discuss the lessons learnt.; Nous notons un besoin crucial d'outils permettant de concevoir des agroécosystèmes durables basés sur l'agroécologie. En effet, de tels agroécosystèmes sont intrinsèquement complexes et leur conception nécessite d'intégrer diverses données et connaissances scientifiques non stabilisées. Dans cette thèse, nous considérons la question de la sélection des espèces de plantes de service en fonction de leur potentiel à fournir des services écosystémiques. Pour aborder cette problématique, nous adoptons une approche basée à la fois sur une représentation formalisée des connaissances du domaine, qui permet de raisonner, et sur l'exploitation des données disponibles, collectées indépendamment de l'application visée. Plus précisément, nous nous appuyons d'une part sur des résultats scientifiques récents en agronomie reliant des traits fonctionnels (i.e., des caractéristiques mesurables des espèces végétales) aux services écosystémiques, et d'autre part sur les données relatives aux traits fonctionnels collectées par la communauté de recherche en écologie. L'architecture de notre système s'inspire du paradigme d'accès aux données basé sur l'ontologie, qui permet de combiner les données et les connaissances de manière raisonnée. Nous fournissons une méthodologie pour acquérir des connaissances scientifiques sous la forme de diagrammes liés à des sources de données, ainsi qu'une formalisation dans un langage logique basé sur des règles. Il est important de noter que nos règles sont indépendantes des diagrammes et des données spécifiques, afin de garantir la généricité et de faciliter l'évolution du système. Nous mettons en œuvre notre méthodologie dans l'étude de cas de l'enherbement des vignes, c'est-à-dire l'installation de plantes de service herbacées dans les vignobles. Nous fournissons une évaluation des résultats du système sur ce cas d'utilisation et discutons des leçons apprises.

Objectives The aim of the study is to design an ontology model for the representation of assets and its features in distributed health care environments. Allow the interchange of information about these assets through the use of specific vocabularies based on the use of ontologies. Methods Ontologies are a formal way to represent knowledge by means of triples composed of a subject, a predicate, and an object. Given the sensitivity of network assets in health care institutions, this work by using an ontology-based representation of information complies with the FAIR principles. Federated queries to the ontology systems, allow users to obtain data from multiple sources (i.e., several hospitals belonging to the same public body). Therefore, this representation makes it possible for network administrators in health care institutions to have a clear understanding of possible threats that may emerge in the network. Results As a result of this work, the “Software Defined Networking Description Language—CUREX Asset Discovery Tool Ontology” (SDNDL-CAO) has been developed. This ontology uses the main concepts in network assets to represent the knowledge extracted from the distributed health care environments: interface, device, port, service, etc. Conclusion The developed SDNDL-CAO ontology allows to represent the aforementioned knowledge about the distributed health care environments. Network administrators of these institutions will benefit as they will be able to monitor emerging threats in real-time, something critical when managing personal medical information.

E-newspaper readers are overloaded with massive texts on e-news articles, and they usually mislead the reader who reads and understands information. Thus, there is an urgent need for a technology that can automatically represent the gist of these e-news articles more quickly. Currently, popular machine learning approaches have greatly improved presentation accuracy compared to traditional methods, but they cannot be accommodated with the contextual information to acquire higher-level abstraction. Recent research efforts in knowledge representation using graph approaches are neither user-driven nor flexible to deviations in the data. Thus, there is a striking concentration on constructing knowledge graphs by combining the background information related to the subjects in text documents. We propose an enhanced representation of a scalable knowledge graph by automatically extracting the information from the corpus of e-news articles and determine whether a knowledge graph can be used as an efficient application in analyzing and generating knowledge representation from the extracted e-news corpus. This knowledge graph consists of a knowledge base built using triples that automatically produce knowledge representation from e-news articles. Inclusively, it has been observed that the proposed knowledge graph generates a comprehensive and precise knowledge representation for the corpus of e-news articles.

This research aims to reveal students' perception-based knowledge representation from mathematics programs in Euclidean Parallelism. Students were asked to do parallelism exercises presented in a verbal and picture form. The data were analyzed by knowledge representation theory based on meaning and perception. There were students who have amodal-multimodal-transition hypothesis. Students' assumptions about the verbal and picture information of not-perfectly-drawn parallel lines varied: assuming that angles appear to be the same measure are congruent, the two lines would intersect, considering the two lines parallel but redrawing picture to determine the pair of congruent angles and using other perspectives to interpret the picture. This study recommends action research for geometry learning that provides not-perfectly-drawn parallel lines for students who have amodal and amodalmultimodal- transition hypothesis and observe its effect on their non-Euclidean geometry learning. It may also familiarize students with getting to know parallelism in R3.

Summary Objectives: We examine the knowledge ecosystem of COVID-19, focusing on clinical knowledge and the role of health informatics as enabling technology. We argue for commitment to the model of a global learning health system to facilitate rapid knowledge translation supporting health care decision making in the face of emerging diseases. Methods and Results: We frame the evolution of knowledge in the COVID-19 crisis in terms of learning theory, and present a view of what has occurred during the pandemic to rapidly derive and share knowledge as an (underdeveloped) instance of a global learning health system. We identify the key role of information technologies for electronic data capture and data sharing, computational modelling, evidence synthesis, and knowledge dissemination. We further highlight gaps in the system and barriers to full realisation of an efficient and effective global learning health system. Conclusions: The need for a global knowledge ecosystem supporting rapid learning from clinical practice has become more apparent than ever during the COVID-19 pandemic. Continued effort to realise the vision of a global learning health system, including establishing effective approaches to data governance and ethics to support the system, is imperative to enable continuous improvement in our clinical care.

Objectives This article presents a comparative study of two Health Level Seven International (HL7) standards for clinical knowledge representation, the Arden Syntax and the Clinical Quality Language (CQL), regarding their expressiveness and utility to represent knowledge for clinical decision support (CDS) systems. Methods We compiled a concatenated set of features from both languages and made descriptive comparisons of 27 categories covering areas of language characteristics, data, control statements, and operators. Results Both Arden and CQL have similar constructs that can be used for representing CDS knowledge but also have unique constructs that could support distinct use cases. They have constructs that fully or partially address several of the categories used in the comparison, except for data models and terminologies in Arden and event triggering and iteration statements in CQL. Conclusion These standards can facilitate the sharing, management, and reuse of computable knowledge, and permit knowledge to be represented with their languages and converted to a machine-friendly executable code that can be shared and reused by other systems. Having support for standard data models and terminologies will continue to be a differential for adoption of a language. The HL7 working groups responsible for developing these standards can direct future development to enhance the functions of the standard and address the gaps identified in this study.

We present a general logical framework for reasoning about agents' cognitive attitudes of both epistemic type and motivational type. We show that it allows us to express a variety of relevant concepts for qualitative decision theory including the concepts of knowledge, belief, strong belief, conditional belief, desire, conditional desire, strong desire and preference. We also present two extensions of the logic, one by the notion of choice and the other by dynamic operators for belief change and desire change, and we apply the former to the analysis of single-stage games under incomplete information. We provide sound and complete axiomatizations for the basic logic and for its two extensions. The paper is under consideration in Theory and Practice of Logic Programming (TPLP)., Under consideration in Theory and Practice of Logic Programming (TPLP)

Narratives provide a powerful means of making sense of our world. They are cognitive tools that we use to examine and shape ourselves, as well as the environment around us. Every day, we seek out and tell stories to reflect on our past, navigate our present, and direct our future. Because narratives embed the complexity of the universe as we know it, it is challenging to develop a complete representation of narrative meaning. Some proposed theories focus on the perceptions and interactions of narrative agents. Others emphasize the communicative impact that a narrative has on its audience. Many more interpretations have been explored, but relatively little work exists that seeks to reconcile these multiple definitions and applications of narrative. Narrative models may reflect more about their creators and their unique scenarios than the elusive tangles of meaning they are meant to capture. They are, after all, stories we are telling about stories. Each presents a lens in which to view, and subsequently define, narrative for a particular context. Regardless, perhaps we can learn from these diverse perspectives. Their congruence illuminates key features of story, while their discrepancies progressively carve out the rich, explorable space of all possible narratives. Although we may never reach the boundaries of that space, our explorations could give rise to new insights in narrative understanding.This thesis seeks to determine whether we can devise a framework that can usefully encode multiple narrative representations in a human-readable format. To do so would promote annotation, comparison, and sharing of narrative data among researchers. Further, a framework that can express distinct narrative representations should also reveal methods for automatically transforming their fundamental units. I will provide examples for how this type of adaptation may be useful for personalizing human-computer interaction scenarios which rely on narrative structures.

The Adaptive User Interface (AUI) adapts to the changes in the context of use and provides improved interaction abilities for different users. The adaptivity in the user interfaces requires in depth knowledge of context. There is a need to enrich user profiles to achieve the personalized services with the ability to adapt the user’s context. The context can be reflected in a particular kind of knowledge and hence modeled as ontology. Ontology based context models are effective means to handle complex situations that support the sharing or integration of context information. This paper presents ontology based context model using OWL for adaptive mobile devices. It models the context over its four major elements including device, user, environment (location and time) and activity. The proposed ontology was derived in different classes, relationships, associations, dependencies and constraints to model dynamic context. Ontologies present a standardized, consistent and shareable context model. The context model and consequent context snapshots can be acknowledged by AUI to present a suitable user interface. The ontology was developed using Protégé on the basis of each context type having different values. Semantic querying (SPARQL) was used for knowledge acquisition. Moreover, the Pellet and HermiT Reasoner were used to verify the rules, relations and constraints to avoid the inconsistency between classes. Comparative to other context models for adaptive interfaces, ontological model provides more of scalability and growth with learning new context in to the shared context knowledge.

Additive Manufacturing (AM) evolved recently from a rapid prototyping process to a standard manufacturing tool. Nevertheless, it is still not a widely used method due to different process-related challenges. In recent years printer technologies and possible printable materials emerged but there are still challenging demands on the printing process. Hence, it is of vital importance to inspect the manufacturability of the designed parts. This work focuses on the not yet widely researched ceramic printing with the Lithography-based Ceramic Manufacturing (LCM) processes. It presents a knowledge-driven framework able to automatically examine geometric properties of a part and compare it to AM guidelines. As a knowledge base, an ontology is used which contains information about the capabilities of AM processes, printers and materials. The manufacturability system uses triangle-based mesh processing algorithms to recognize features and check the guidelines necessary for LCM. The evaluation shows the feasibility of manufacturability analysis with the developed framework and its limitations.

Communication au congrès KES de 2021 qui s'est déroulé à Szczecin; International audience; Understanding the replication machinery of viruses contributes to suggest and try effective antiviral strategies. Exhaustive knowledge about the proteins structure, their function, or their interaction is one of the preconditions for successfully modeling it. In this context, modeling methods based on a formal representation with a high semantic expressiveness would be relevant to extract proteins and their nucleotide or amino acid sequences as an element from the replication process. Consequently, our approach relies on the use of semantic technologies to design the SARS-CoV-2 replication machinery. This provides the ability to infer new knowledge related to each step of the virus replication. More specifically, we developed an ontology-based approach enriched with reasoning process of a complete replication machinery process for SARS-CoV-2. We present in this paper a partial overview of our ontology OntoRepliCov to describe one step of this process, namely, the continuous translation or protein synthesis, through classes, properties, axioms, and SWRL (Semantic Web Rule Language) rules.

One of the critical aspects of building intelligent tutoring systems regards proper monitoring of student's activity and academic performance. This paper presents a continuous student knowledge tracing method implemented for Tesys e-Learning platform at the Faculty of Automation, Computers and Electronics in the University of Craiova. The student's knowledge level is continuously monitored and, after each recommended test by the SVD-based mechanism, a new set of knowledge weights are computed. We aim to achieve a comprehensive monitoring environment which can provide an accurate insight upon the student's knowledge level at any moment. In our approach, we added weights for both students and tests to improve the student's evolution monitoring and provide more accurate feedback. The setup for validation consisted of ten tests with eight questions per test and we used both current and past year tests data. Results revealed that assigning weights to questions, tests and students and using them in the recommendation process offers a better view of the student's evolution along with more accurate recommendations. Progress in this direction will provide more insight into available teaching materials and SVD-based recommender system such that the e-learning platform that integrates the presented mechanism will provide a better learning experience.

Knowledge generated from appliances usage data can be used for personalizing and reducing the number of user interactions. Appliances such as Washing Machines (WM) are capable of storing the settings data selected by the user, during operation. Before the operation of a WM, the user has to enable the options for these settings for a preferred mode of operation. We have proposed a Device Interaction Graph solution that automatically recommends and auto-completes the user's preferred settings in real time. For direct manual interactions on the WM, the auto-completion solution is applied when the user incrementally changes the settings on the user interface. For voice based command interactions with the WM, the auto-completion solution would be applied after understanding any incomplete settings requirements mentioned by the user. This settings auto-completion solution improves the user experience by reducing the number of manual or verbal interactions required to enable the operation of the device. We have applied the auto-completion solution on the data obtained from 158,213 connected WMs and achieved recommendation accuracy of 89.79% with an average 3.132 interactions.

The "Node of Knowledge" method is one of the elements of the conceptual framework "Node of Knowledge (NOK)". It enables knowledge representation in graphical and formalized (textual) form and makes it possible to store formalized records of natural language sentences in a relational database. To enable correct transformation of all words from natural language sentences to formalized records, it is necessary to design a metamodel of a language, i.e. to analyse all word types of each particular natural language and define rules for transformation of sentences in natural language into formalized records. This paper analyses nouns in Croatian and English language. It presents rules for transformation of nouns and noun phrase structures into formalized records, with examples in both languages. The system is preliminary tested with a small set of sentences (used as input knowledge) and questions. Testing results are presented and discussed.

Credibility reasoning has attracted a lot of attention due to its distinguished power and efficiency in representing uncertainty and vagueness within the process of reasoning and decision making. Aiming at the problem of inaccurate credibility estimation in uncertainty reasoning and making experts to express hesitant preferences better in evaluation reasoning process, this paper introduces hesitant fuzzy linguistic term set into credibility uncertainty reasoning. First, we propose hesitant fuzzy linguistic-valued credibility (HLCF), and establish the knowledge representation model of the hesitant fuzzy linguistic-valued credibility. Then, in order to solve the problem of incomplete information in the evaluation reasoning process, an information complement algorithm based on maximum similarity is constructed. After that, the algorithms of single rule and multiple rules of parallel relationship of hesitant fuzzy linguistic-valued credibility are proposed to enrich the reasoning rule base and get more accurate reasoning results. The closeness degrees between the conclusions of each alternative after reasoning and the expected value are calculated, so as to select the most suitable alternative. Finally, a practical example which concerned the social risk analysis is given to illustrate the applicability and effectiveness of the proposed approach.

Graph embedding models are widely used in knowledge graph completion (KGC) task. However, most models are based on the assumption that knowledge is completely certain, and this is inconsistent with real-world situations. Although there are multiple studies on uncertain knowledge embedding tasks, they often use knowledge confidence to learn embedding and cannot make full use the structural and uncertain information of knowledge. This paper presents a new embedding model named Structural and Uncertain Knowledge Embedding (SUKE), which comprises two components: an evaluator and a confidence generator. For unknown triples, the evaluator learns the structural and uncertain information to evaluate its rationality and obtain a candidate set. The confidence generator then determines the confidence of the candidate set to achieve KGC. To verify the effectiveness of the proposed model, confidence prediction, triple evaluation, and fact classification tasks are performed on three data sets. Experimental results show that SUKE performs better than mainstream embedding methods. The model proposed in this paper can help advance the research on the embedding of uncertain knowledge graphs.

The scientific novelty of the results is as follows. A temporal approach to constructing explanations for the operation of an intelligent system is proposed. The approach describes explanation as a process consisting of a temporally ordered sequence of facts. The order of time for pairs of facts is determined by temporal rules. Such rules may define the explanation process with varying degrees of detail over time, depending on the request for clarification. Detailed explanations reflect the subject area model and include the basic and alternative sequences of actions performed by the intelligent system. The explanation of the basic patterns of the intelligent system makes it possible to interpret the limitations that affect the obt ained solution. The explanation of the system as a whole provides an implicit reflection of the key causal relationships, which allows you to get a simplified interpretation of the results of the intelligent system. A dynamic model of describing explanations based on temporal knowledge for use in the human-computer interface is proposed. The model takes into account the description of actions in the subject area, the patterns of these actions, as well as generalized causal relationships between such patterns. The model provides an opportunity to present the dynamics of the process of functioning of the intelligent system with the required level of detail, as well as change the level of detail to clarify the explanation at the request of the user. Наукова новизна отриманих результатів полягає в наступному. Запропоновано темпоральний підхід до побудови пояснень щодо роботи інтелектуальної системи. Підхід описує пояснення як процес, що складається із темпорально упорядкованої послідовності фактів. Порядок у часі для пар фактів задається темпоральнимиправилами. Такі правила можуть визначати процес пояснення із різним ступенем деталізації у часі, в залежності від запиту на пояснення. Детальні пояснення відображають модель предметної області і містять у собі базові та альтернативні послідовності дій, що виконує інтелектуальна система. Пояснення базових патернів роботи інтелектуальної системи дає можливість інтерпретувати обмеження, які впливають на отримане рішення. Пояснення роботи системи в цілому забезпечує неявне відображення ключових причинно-наслідкових залежностей, що дозволяє отримати спрощену інтерпретацію результатів роботи інтелектуальної системи. Запропоновано динамічну модель опису пояснень на основі темпоральних знань для використання у людино-машинному інтерфейсі. Модель враховує опис дій у предметної області, патерни цих дій, а також узагальнені причинно-наслідкових зв'язки між такими патернами. Модель забезпечує можливість представлення динаміки процесу функціонування інтелектуальної системи із необхідним ступенем деталізації, а також зміну рівня деталізації для уточнення пояснення за запитом користувача.