Showing papers presented at "International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management in 2020"

CODO: An Ontology for Collection and Analysis of Covid-19 Data

Abstract: The COviD-19 Ontology for cases and patient information (CODO) provides a model for the collection and analysis of data about the COVID-19 pandemic. The ontology provides a standards-based open-source model that facilitates the integration of data from heterogeneous data sources. The ontology was designed by analysing disparate COVID-19 data sources such as datasets, literature, services, etc. The ontology follows the best practices for vocabularies by re-using concepts from other leading vocabularies and by using the W3C standards RDF, OWL, SWRL, and SPARQL. The ontology already has one independent user and has incorporated real-world data from the government of India.

Some Techniques for Intelligent Searching on Ontology-based Knowledge Domain in e-Learning.

Abstract: E-learning is the modern way to learn by using electronic media and information and communication technologies in education. Ontology is a useful method to organize knowledge bases for intelligent educational systems. In this paper, a method for intelligent searching on ontology-based knowledge domain in e-learning is presented. This method includes an ontology representing educational knowledge domains, called Search-Onto. The foundation of this ontology is concepts, relations between concepts, operators, and rules combining the structures of problems and their solving methods. Beside the ontology organizing the knowledge base, the proposed method also studies some techniques for intelligent searching, such as searching for the knowledge content, searching on the knowledge classification, and searching the related knowledge. The method for intelligent searching based on a knowledge base has been applied to construct a search engine for the knowledge of high-school mathematics. This engine can do searching works and retrieve required information in mathematics for high-school students to support their learning.

Historical Document Processing: A Survey of Techniques, Tools, and Trends.

Abstract: Historical Document Processing (HDP) is the process of digitizing written material from the past for future use by historians and other scholars. It incorporates algorithms and software tools from computer vision, document analysis and recognition, natural language processing, and machine learning to convert images of ancient manuscripts and early printed texts into a digital format usable in data mining and information retrieval systems. As libraries and other cultural heritage institutions have scanned their historical document archives, the need to transcribe the full text from these collections has become acute. Since HDP encompasses multiple sub-domains of computer science, knowledge relevant to its purpose is scattered across numerous journals and conference proceedings. This paper surveys the major phases of HDP, discussing standard algorithms, tools, and datasets and finally suggests directions for further research.

A recurrent neural network and differential equation based spatiotemporal infectious disease model with application to covid-19

Abstract: The outbreaks of Coronavirus Disease 2019 (COVID-19) have impacted the world significantly. Modeling the trend of infection and real-time forecasting of cases can help decision making and control of the disease spread. However, data-driven methods such as recurrent neural networks (RNN) can perform poorly due to limited daily samples in time. In this work, we develop an integrated spatiotemporal model based on the epidemic differential equations (SIR) and RNN. The former after simplification and discretization is a compact model of temporal infection trend of a region while the latter models the effect of nearest neighboring regions. The latter captures latent spatial information. We trained and tested our model on COVID-19 data in Italy, and show that it out-performs existing temporal models (fully connected NN, SIR, ARIMA) in 1-day, 3-day, and 1-week ahead forecasting especially in the regime of limited training data. Copyright © 2020 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.

SatelliteNER: An Effective Named Entity Recognition Model for the Satellite Domain.

Abstract: Named Entity Recognition (NER) is an important task that detects special type of entities in a given text. Existing NER techniques are optimized to find commonly used entities such as person or organization names. They are not specifically designed to find custom entities. In this paper, we present an end-to-end framework, called SatelliteNER, that its objective is to specifically find entities in the Satellite domain. The workflow of our proposed framework can be further generalized to different domains. The design of SatelliteNER includes effective modules for preprocessing, auto-labeling and collection of training data. We present a detailed analysis and show that the performance of SatelliteNER is superior to the state-of-the-art NER techniques for detecting entities in the Satellite domain.

Digitalization: A Concept Easier to Talk about than to Understand.

Abstract: Many organizations have come to rely on digitalization to solve many issues. Today knowledge is an equal production factor besides the traditional ones; capital, natural resources, and work. Thus, there is an evergrowing need for getting the information in, sorted, and used. Digitalization is widely used phrase with many definitions, quite often case-specifically. We show that the existing definitions are not very precise and through two studies investigating executed digitalization initiatives we point out that the reality does not always respect earlier findings. Through the comparison we present questions to be answered in later research. However, business environments and technologies are often unique and thus not all recent issues are taken into consideration. We take business intelligence (BI) as a framework to draw a picture of organizational processes and to give context for the features to be taken into account when discussing digitalization, ie. technological side and the human oriented aspects.

Semantic Representation of Physics Research Data.

Abstract: Improvements in web technologies and artificial intelligence enable novel, more data-driven research practices for scientists. However, scientific knowledge generated from data-intensive research practices is disseminated with unstructured formats, thus hindering the scholarly communication in various respects. The traditional document-based representation of scholarly information hampers the reusability of research contributions. To address this concern, we developed the Physics Ontology (PhySci) to represent physics-related scholarly data in a machine-interpretable format. PhySci facilitates knowledge exploration, comparison, and organization of such data by representing it as knowledge graphs. It establishes a unique conceptualization to increase the visibility and accessibility to the digital content of physics publications. We present the iterative design principles by outlining a methodology for its development and applying three different evaluation approaches: data-driven and criteria-based evaluation, as well as ontology testing.

CoExDBSCAN: Density-based Clustering with Constrained Expansion –

Abstract: Full space clustering methods suffer the curse of dimensionality, for example points tend to become equidistant from one another as the dimensionality increases. Subspace clustering and correlation clustering algorithms overcome these issues, but still face challenges when data points have complex relations or clusters overlap. In these cases, clustering with constraints can improve the clustering results, by including a priori knowledge into the clustering process. This article proposes a new clustering algorithm CoExDBSCAN, density-based clustering with constrained expansion, which combines traditional, density-based clustering with techniques from subspace, correlation and constrained clustering. The proposed algorithm uses DBSCAN to find densityconnected clusters in a defined subspace of features and restricts the expansion of clusters to a priori constraints. We provide verification and runtime analysis of the algorithm on a synthetic dataset and experimental evaluation on a climatology dataset of satellite observations. The experimental dataset demonstrates, that our algorithm is especially suited for spatio-temporal data, where one subspace of features defines the spatial extent of the data and another correlations between features.

Connecting the Dots: KM Initiatives and Business Performance.

Abstract: For over two decades now, knowledge management (KM) has been an academic discipline, extensively taught, learned, and researched. It has been practiced in organizations for a similar period. However, demonstrating long-term improvement in business performance as a result of KM initiatives is not an easy task. KM processes and changes are rarely standalone and usually go hand in hand with other organizational changes. This in itself makes it difficult to demonstrate that said organizational changes originate from KM initiatives, and not from other changes in the organizational environment. This research provides a case study in which shifts in organizational performance as a result of KM initiatives can be examined. This research is significant for KM researchers because it suggests a new research model for further studies. In addition, it provides organizations with an optimistic vision, namely that systematic management of organizational knowledge assets is not merely an idea that sounds good, but can actually be proven to be effective. Connecting the dots between KM and business performance is a known challenge. However, thanks to our research model, it is easier to overcome this challenge. The research method is based on a qualitative and quantitative case study in which foster care activities in Israel underwent KM intervention for a period of eight years. During this period, failures of foster care cases resulting in unsuccessful placement dropped by 32%, with KM being a major factor contributing to this

Automatic ontology learning from domain-specific short unstructured text data

Abstract: Ontology learning is a critical task in industry, dealing with identifying and extracting concepts captured in text data such that these concepts can be used in different tasks, e.g. information retrieval. Ontology learning is non-trivial due to several reasons with limited amount of prior research work that automatically learns a domain specific ontology from data. In our work, we propose a two-stage classification system to automatically learn an ontology from unstructured text data. We first collect candidate concepts, which are classified into concepts and irrelevant collocates by our first classifier. The concepts from the first classifier are further classified by the second classifier into different concept types. The proposed system is deployed as a prototype at a company and its performance is validated by using complaint and repair verbatim data collected in automotive industry from different data sources.

Towards Construction of Legal Ontology for Korean Legislation.

Abstract: Automating information extraction from legal documents and formalising them into a machine understandable format has long been an integral challenge to legal reasoning. Most approaches in the past consist of highly complex solutions that use annotated syntactic structures and grammar to distil rules. The current research trend is to utilise state-of-the-art natural language processing (NLP) approaches to automate these tasks, with minimum human interference. In this paper, based on its functional aspects, we propose a legal taxonomy of semantic types in Korean legislation, such as definitional provision, deeming provision, penalty, obligation, permission, prohibition, etc. In addition to this, a NLP classifier has been developed to facilitate the automated legal norms classification process and an overall F1 score of 0.97 has been achieved.

Digital Transformation and the Impact in Knowledge Management.

Abstract: Nowadays, digital transformation is forcing companies to reach a new level of productivity and digital evolution. Small and autonomous is winning over large and centralized. Digital transformation requires the adoption of more agile business processes and the development of new customer-facing digital services. It means creating scale through reusable services and enabling self-service consumption of those services. Business processes and transactions can be automated with the composition of microservices. We will see that the principle of composability allows microservices to deliver value to the business in different contexts. The paper also explains how a BizDevOps philosophy with references to microservices allows rapid adaptations of requirements to fast-changing needs in businesses, outlining the importance of business process automation for companies to acquire the know-how to implement a just-in-time diachronic dialogue. It presents the alignment of the proposed framework with a digital strategy. Assembling a multidisciplinary team is foreseen as a key factor in developing innovative capabilities to react to new customer demands, enabling the company to stay competitive and continuously address customer expectations, differentiating tacit from explicit knowledge.

An Ontology of Chinese Ceramic Vases.

Abstract: Extensive collections of Chinese ceramic vases are housed in museums throughout China. They could serve as rich sources of data for historical research. Although some data sources have been digitized, the vision of heritage institutions is not only to display objects and simple descriptions (drawn from metadata) but also to allow for understanding relationships between objects (created by semantically interrelated metadata). The key to achieving this goal is to utilize the technologies of the Semantic Web, whose core is Ontology. The focus of this paper is to describe the construction of the TAO CI (“ceramics” in Chinese) ontology of the domain of ceramic vases of the Ming (1368-1644) and Qing (1644-1911) dynasties. The theoretical and methodological approach adopted to construct the TAO CI ontology is term-and-characteristic guided, i.e., it relies on a morphological analysis of the Chinese terms used in the domain, and respects the ISO principles on Terminology (ISO 1087 and 704), according to which concepts are defined by means of essential characteristics. The research presented in this article aims to publish the resulting structured data on the Semantic Web for the use of anybody interested, including museums hosting collections of these vessels, and to enrich existing methodologies on domain ontology building. To our knowledge, there are no comprehensive ontologies for Chinese ceramic vases. TAO CI ontology remedies this gap and provides a reference for ontology building in other domains of Chinese cultural heritage. The tool used is Protégé. The TAO CI ontology is open access here: http://www.dh.ketrc.com/otcontainer/data/OTContainer.owl.

Next Step: Data Literacy Measurement.

Abstract: As data became a new business commodity, affecting our everyday lives from shopping to voting, it smoothed the way for data literacy as a tool for full participation in a modern society. This paper argues for data literacy development and accelerated research of its measurement which has been lagging behind countless studies on teaching data skills. Data literacy is in this paper approached as an ability to understand and use data and differentiates itself from information or statistical literacy. As a prerequisite of information literacy, data literacy is inevitable part of knowledge development. While the term of data literacy has been well established and used for developing best practices and methodologies to teach data skills, measurement of data literacy seems to be still in its infancy. As a result, this paper includes research plan for developing a data literacy indicator based on quantitative methods.

Ontology-based Processing of Dynamic Maps in Automated Driving.

Abstract: Autonomous cars act in a highly dynamic environment and consistently have to provide safety and comfort to the passengers. For a car to understand its surroundings, a detailed, high-definition digital map is needed, which acts as a powerful virtual “sensor”. Compared to traditional digital maps, high-definition maps require significantly more storage space, which makes it largely impossible to store a complete map in a navigation system. Furthermore, map data is provided in numerous heterogeneous formats. Consequently, interoperability and scalability have become the main challenges of existing map processing solutions. We address these challenges by providing an interoperable knowledge-spatial architecture layer based on ontologies and confirm the scalability in an empirical evaluation.

The Max-Cut Decision Tree: Improving on the Accuracy and Running Time of Decision Trees.

Abstract: Decision trees are a widely used method for classification, both by themselves and as the building blocks of multiple different ensemble learning methods. The Max-Cut decision tree involves novel modifications to a standard, baseline model of classification decision tree construction, precisely CART Gini. One modification involves an alternative splitting metric, maximum cut, based on maximizing the distance between all pairs of observations belonging to separate classes and separate sides of the threshold value. The other modification is to select the decision feature from a linear combination of the input features constructed using Principal Component Analysis (PCA) locally at each node. Our experiments show that this node-based localized PCA with the novel splitting modification can dramatically improve classification, while also significantly decreasing computational time compared to the baseline decision tree. Moreover, our results are most significant when evaluated on data sets with higher dimensions, or more classes; which, for the example data set CIFAR-100, enable a 49% improvement in accuracy while reducing CPU time by 94%. These introduced modifications dramatically advance the capabilities of decision trees for difficult classification tasks.

Challenges in Public Participation and Collaboration: A Case Study in Finnish Environmental Decision-making.

Abstract: This position paper introduces ongoing research efforts that addresses the ability of different kinds of organizations and multiple individuals to cope together with complex environmental planning and policymaking problems in the Finnish context. The research question “What kind of challenges are there in the collaborative processes of environmental decision-making and how can they be tackled?” is approached from the perspectives of the framework of public participation process and the theory of collaborative governance. We use these theories as analytical tools to evaluate how the elements and phases of collaboration processes are conducted in practice and to identify problems that exist in the collaborative processes. This phenomenon is studied through a single case study of environmental planning case from a medium-sized city located in

Persuasion Meets AI: Ethical Considerations for the Design of Social Engineering Countermeasures.

Abstract: Privacy in Social Network Sites (SNSs) like Facebook or Instagram is closely related to people's self-disclosure decisions and their ability to foresee the consequences of sharing personal information with large and diverse audiences. Nonetheless, online privacy decisions are often based on spurious risk judgements that make people liable to reveal sensitive data to untrusted recipients and become victims of social engineering attacks. Artificial Intelligence (AI) in combination with persuasive mechanisms like nudging is a promising approach for promoting preventative privacy behaviour among the users of SNSs. Nevertheless, combining behavioural interventions with high levels of personalization can be a potential threat to people's agency and autonomy even when applied to the design of social engineering countermeasures. This paper elaborates on the ethical challenges that nudging mechanisms can introduce to the development of AI-based countermeasures, particularly to those addressing unsafe self-disclosure practices in SNSs. Overall, it endorses the elaboration of personalized risk awareness solutions as i) an ethical approach to counteract social engineering, and ii) as an effective means for promoting reflective privacy decisions.

Annotating Arguments in a Parliamentary Corpus: An Experience.

Abstract: Estonian parliamentary corpus includes verbatim records of sessions held in the Parliament of Estonia (Riigikogu) in 1995-2001. An important task of the Riigikogu is the passing of acts and resolutions. A bill initiated in the Riigikogu will pass three readings, during which it is refined and amended. Negotiation is an important part of parliamentary discussions. Arguments for and against of the bill and its amendments are presented by the members of the Parliament in negotiation. In the paper, arguments used in negotiation are considered. Every argument consists of one or more premises, and a claim (or conclusion). The arguments and the relations between them (rebuttal, attack, and support) are determined with the aim to create a corpus where arguments are annotated. Some problems are discussed in relation with annotation. Our further aim is the automatic recognition of arguments and inter-argument relations in Estonian political texts.

Driving the Bus: A Radiology Case Study Utilizing Rich Picture Diagramming and CATWOE to Capture Staff Perceptions and Inform Service Improvement Scenarios.

Abstract: Introduction: Demand for computed tomography (CT) services is increasing. This empirical work in a CT setting, examines how qualitative methods were used to identify staff perceptions and opportunities for service improvement. The use of soft systems methodologies (SSM) as an action research tool in radiology is considered. Methods: Hospital Staff were interviewed to create a root definition of the CT service. In a diagramming session, a rich picture (RP) was created and augmented with staff input. Utilizing the RP, a session was facilitated with radiology decision makers to identify a list of culturally desirable and feasible process improvement scenarios. Results: Root definitions were created of the CT service from the perspective of the staff. The RP graphically illustrated the key features of the CT service and represented a shared understanding of the service. A targeted set of culturally feasible and desirable service improvement recommendations were identified. Four directly attributable implemented workflow changes were identified. Conclusion: RP diagramming provided an opportunity to involve staff in research while capturing their perceptions and resulted in a shared understanding as well as targeted opportunities for CT service improvement. The implemented workflow changes resulting from the SSM approach demonstrated its use as an action research tool. a https://orcid.org/0000-0001-7241-8869 b https://orcid.org/0000-0002-8781-9742

Collaborative Virtual Reality as an Adaptable Boundary Object in the Design Phase of Facility Life Cycle

Abstract: Although the large majority of costs of buildings incur in the later operation and maintenance phase, major decisions affecting these costs are made in the early design and construction phases. Virtual Reality (VR) and Collaborative Virtual Reality (CVR) have been noticed to have significant potential in involving the expertise and needs of various stakeholders into the early design phases, increasing the quality of building designs and reducing related costs. Boundary Object Theory has been noticed useful in better understanding and improving the knowledge transfer of actors with different backgrounds and expertise. VR and CVR remain yet little studied as boundary objects. We will address this research gap in this study by aiming to understand how CVR can act as an adaptable boundary object in the building design phase of the facility life cycle. We have made use of a qualitative approach, consisting of a multiple case study approach and semi-structured interviews in Finnish AEC industry companies and organisations. We contribute to academic research by providing a deeper understanding of how CVR functions as a boundary object, which enhances the transfer of knowledge in new ways between various stakeholders in the building design phase.

Session similarity based approach for alleviating cold-start session problem in e-commerce for Top-N recommendations

Abstract: Cold-start problem is one of the main challenges for the recommender systems. There are many methods developed for traditional recommender systems to alleviate the drawback of cold-start user and item problems. However, to the best of our knowledge, in session based recommender systems cold-start session problem still needs to be investigated. In this paper, we propose a session similarity-based method to alleviate drawback of cold-start sessions in e-commerce domain, in which there are no interacted items in the sessions that can help to identify users’ preferences. In the proposed method, product recommendations are given based on the most similar sessions that are found using session features such as session start time, location, etc. Computational experiments on two real-world datasets show that when the proposed method applied, there is a significant improvement on the performance of recommender systems in terms of recall and precision metrics comparing to random recommendations for cold-start sessions.

Evaluating Task Knowledge as a Mediator in the Relationship between Knowledge Sharing and Innovative Work Behaviour.

Abstract: It is widely recognized that knowledge sharing contributes to innovation in organizations. The implicit assumption in the linkage between individual knowledge sharing and their innovative work behaviour is that individuals gain certain qualities while being engaged in knowledge sharing that enable them to become more innovative. In this paper, we explore employee task knowledge as a key mediator in the relationship between knowledge sharing and their innovative work behaviour. Data collected from knowledge workers from several manufacturing and service based organizations is used to test the mediation hypotheses. Results support our mediation hypothesis and show that task knowledge partially mediates knowledge sharing’s impact on innovative work behaviour. Knowledge sharing had a positive impact on innovative work behaviour even after considering task knowledge as a mediator, suggesting other mechanisms at work in addition to their task knowledge in how knowledge sharing contributes to innovation. Theoretical and practical implications of the findings are also discussed.