Search

In this paper, we derived competences from previously developed competence models, ensuring the effective use of advanced technologies in future factories to improve the sustainability of their business models and strategies. Based on the analysis of the Hogan competence model and competence models for sustainability and leadership, we compiled a selection of competences for digitalisation, automation, robotics, artificial intelligence, and soft competences such as emotional intelligence and cultural literacy. We also included competences required for sustainability, corporate social responsibility, and circular economy. The selected competences formed the core for the conceptual development of a decision support tool for the individualised selection of training for employees. The concept was tested in customised training to improve employees’ skills and motivation for lifelong learning at the selected industrial partner. The developed assessment algorithm was used to monitor the progress of individual employees’ skills development before and after their training participation. The results of the assessment help human resource departments make decisions for selecting the most effective and optimal training for employees to improve their sustainability-related competences. Such a systematic approach can improve and evaluate competences that companies need to transition to a circular economy.

The field of bisociative literature-based discovery aims at mining scientific literature to reveal yet uncovered connections between different fields of specialization. This paper outlines several outlier-based literature mining approaches to bridging term detection and the lessons learned from selected biomedical literature-based discovery applications. The paper addresses also new prospects in bisociative literature-based discovery, proposing an advanced embeddings-based technology for cross-domain literature mining.

A result-oriented engagement system for performance optimisation (RESPO) has been developed to systematically monitor and improve the competencies of individuals in business, lifelong learning and secondary schools. The RESPO expert system was transferred for use in higher education institutions (HEIs) based on successful practical application trials. The architecture and functionality of the original RESPO expert system have been transformed into a new format that will collect information on the required competencies and the available educational programmes to help students effectively develop competencies through formal and non-formal education. First, the initial version of the RESPO system and its functionality were tested on a selected group of students and higher education staff to validate and improve its effectiveness for the needs of HEIs. This paper summarises the key findings and recommendations of the validators for transforming the RESPO application into an application for HEIs. In addition, the selection of competencies in the RESPO application database has been adapted to align with selected study programmes and the need to develop sustainability-related competencies. These findings can support professionals working in higher education institutions in developing students’ future competencies and fostering the targeted use of learning analytics tools.

Estimation of probabilities from empirical data samples has drawn close attention in the scientific community and has been identified as a crucial phase in many machine learning and knowledge discovery research projects and applications. In addition to trivial and straightforward estimation with relative frequency, more elaborated probability estimation methods from small samples were proposed and applied in practice (e.g., Laplace’s rule, the m-estimate). Piegat and Landowski (2012) proposed a novel probability estimation method from small samples Ep h√2 that is optimal according to the mean absolute error of the estimation result. In this paper we show that, even though the articulation of Piegat’s formula seems different, it is in fact a special case of the m-estimate, where p a =1/2 and m = √2 . In the context of an experimental framework, we present an in-depth analysis of several probability estimation methods with respect to their mean absolute errors and demonstrate their potential advantages and disadvantages. We extend the analysis from single instance samples to samples with a moderate number of instances. We define small samples for the purpose of estimating probabilities as samples containing either less than four successes or less than four failures and justify the definition by analysing probability estimation errors on various sample sizes.

The aim of the presented human resource monitoring system is to assess performance and provide information necessary for the planning and management of primary health-care network in Slovenia. At the general level, the approach is based on a carefuly designed hierarchy of modules, each monitoring a specific aspect of the network ? in particular physicians' age, qualifications, workload and dispersion. At the implementation level, the system uses a number of different techniques, including OLAP, KDD and data analysis (such as association rule mining). In most cases, the results are presented visually by charts, network graphs and maps. In this way, the monitoring system provides an information-rich picture of the network and its performance, and also helps detecting its critical aspects that require short- or long-term management actions. In principle, the higher levels of the model provide holistic information, while the lower levels provide more details that are useful for the explanation of observed phenomena. The monitoring system has been developed in collaboration with the Ministry of Health of the Republic of Slovenia. Currently, it is implemented as a prototype and has been tested with real data for the year 2006. For the future, we wish that it becomes a regular tool for monitoring the health-care network in Slovenia. We also envision the application of the same methodology in other public networks, such as education and police.

The abundance of literature related to the widespread COVID-19 pandemic is beyond manual inspection of a single expert. Development of systems, capable of automatically processing tens of thousands of scientific publications with the aim to enrich existing empirical evidence with literature-based associations is challenging and relevant. We propose a system for contextualization of empirical expression data by approximating relations between entities, for which representations were learned from one of the largest COVID-19-related literature corpora. In order to exploit a larger scientific context by transfer learning, we propose a novel embedding generation technique that leverages SciBERT language model pretrained on a large multi-domain corpus of scientific publications and fine-tuned for domain adaptation on the CORD-19 dataset. The conducted manual evaluation by the medical expert and the quantitative evaluation based on therapy targets identified in the related work suggest that the proposed method can be successfully employed for COVID-19 therapy target discovery and that it outperforms the baseline FastText method by a large margin., Accepted to the 23rd International Conference on Discovery Science (DS 2020)

In knowledge discovery, experts frequently need to combine knowledge from different domains to get new insights and derive new conclusions. Intelligent systems should support the experts in the search for relationships between concepts from different domains, where huge amounts of possible combinations require the systems to be efficient but also sufficiently general, open and interactive to enable the experts to creatively guide the discovery process. The paper proposes a cross-domain literature mining methodology that achieves this functionality by combining the functionality of two complementary text mining tools: clustering and topic ontology creation tool OntoGen and cross-domain bridging terms exploration tool CrossBee. Focusing on outlier documents identified by OntoGen contributes to the efficiency, while CrossBee allows for flexible and user-friendly bridging concepts exploration and identification. The proposed approach, which is domain independent and can support cross-domain knowledge discovery in any field of science, is illustrated on a biomedical case study dealing with Alzheimer’s disease, one of the most threatening age-related diseases, deteriorating lives of numerous individuals and challenging the ageing society as a whole. By applying the proposed methodology to Alzheimer’s disease and gut microbiota PubMed articles, we have identified Nitric oxide synthase (NOS) as a potentially valuable link between these two domains. The results support the hypothesis of neuroinflammatory nature of Alzheimer’s disease, and is indicative for the quest for identifying strategies to control nitric oxide-associated pathways in the periphery and in the brain. By addressing common mediators of inflammation using literature-based discovery, we have succeeded to uncover previously unidentified molecular links between Alzheimer’s disease and gut microbiota with a multi-target therapeutic potential.

Text mining and natural language processing are fast growing areas of research, with numerous applications in business, science and creative industries. This paper presents TextFlows, a web-based text mining and natural language processing platform supporting workflow construction, sharing and execution. The platform enables visual construction of text mining workflows through a web browser, and the execution of the constructed workflows on a processing cloud. This makes TextFlows an adaptable infrastructure for the construction and sharing of text processing workflows, which can be reused in various applications. The paper presents the implemented text mining and language processing modules, and describes some precomposed workflows. Their features are demonstrated on three use cases: comparison of document classifiers and of different part-of-speech taggers on a text categorization problem, and outlier detection in document corpora.

We improved wordification methodology and provide a formal framework and pseudo code.We statistically evaluated comparable algorithms on multiple relational databases.Experiments show favorable results in terms of accuracy and efficiency.Feature simplicity is compensated by n-gram construction and by feature weighting.We implemented the full experimental workflow in a data mining platform ClowdFlows. Inductive Logic Programming (ILP) and Relational Data Mining (RDM) address the task of inducing models or patterns from multi-relational data. One of the established approaches to RDM is propositionalization, characterized by transforming a relational database into a single-table representation. This paper presents a propositionalization technique called wordification which can be seen as a transformation of a relational database into a corpus of text documents. Wordification constructs simple, easy to understand features, acting as words in the transformed Bag-Of-Words representation. This paper presents the wordification methodology, together with an experimental comparison of several propositionalization approaches on seven relational datasets. The main advantages of the approach are: simple implementation, accuracy comparable to competitive methods, and greater scalability, as it performs several times faster on all experimental databases. Furthermore, the wordification methodology and the evaluation procedure are implemented as executable workflows in the web-based data mining platform ClowdFlows. The implemented workflows include also several other ILP and RDM algorithms, as well as the utility components that were added to the platform to enable access to these techniques to a wider research audience.

Part 1: E-Government Foundations; International audience; To conduct their business, organizations are nowadays challenged to handle huge amount of information from heterogeneous sources. Novel technologies can help them dealing with this delicate assignment. In this paper we describe an approach to document clustering and outlier detection that is regularly used to organize and summarize knowledge stored in huge amounts of documents in a government organization. The motivation for our preliminary study has been three-fold: first, to obtain an overview of the topics addressed in the recently published e-government papers, with the emphasis on identifying the shift of focus through the years; second, to form a collection of papers related to a preselected terms of interest in order to explore the characteristic keywords that discriminate this collection with respect to the rest of the documents; and third, to compare the papers that address a similar topic from two document sources and to show characteristic similarities and differences between the two origins, with a particular aim to identify outlier papers in each document source that are potentially worth for further exploration. As a document source for our study we used E-Government Reference Library of articles and PubMed. The presented case study results suggest that the document exploration supported by a document clustering tool can be more focused, efficient and effective.

To support biomedical experts in their knowledge discovery process, we have developed a literature mining method called RaJoLink for identification of relations between biomedical concepts in disconnected sets of articles. The method implements Swanson’s ABC model approach for generating hypotheses in a new way. The main novelty is a semi-automated suggestion of candidates for agents a that might be logically connected with a given phenomenon c under investigation. The choice of candidates for a is based on rare terms identified in the literature on c. As rare terms are not part of the typical range of information, which describe the phenomenon under investigation, such information might be considered as unusual observations about the phenomenon c. If literatures on these rare terms have an interesting term in common, this joint term is declared as a candidate for a. Linking terms b between literature on a and literature on c are then searched for in the closed discovery to provide additional supportive evidence for uncovered connections. We have applied the method to the literature on autism and have used MEDLINE as a source of data. Expert evaluation has confirmed that the discovered relations might contribute to a better understanding of autism.

This paper presents ways to use subgroup discovery to generate actionable knowledge for decision support. Actionable knowledge is explicit symbolic knowledge, typically presented in the form of rules, that allows the decision maker to recognize some important relations and to perform an appropriate action, such as targeting a direct marketing campaign, or planning a population screening campaign aimed at detecting individuals with high disease risk. Different subgroup discovery approaches are outlined, and their advantages over using standard classification rule learning are discussed. Three case studies, a medical and two marketing ones, are used to present the lessons learned in solving problems requiring actionable knowledge generation for decision support.

Literature-based discovery tools have been often used to overcome the problem of fragmentation of science and to assist researchers in their process of cross-domain knowledge discovery. In this paper we propose a methodology for cross-domain literature-based discovery that focuses on outlier documents to reduce the search space of potential cross-domain links and to improve search efficiency. In a previous study, literature mining tools OntoGen for document clustering and CrossBee for cross-domain bridging term exploration were combined to search for hidden relations in scientific papers from two different domains of interest, where the utility of the approach was demonstrated in a study involving PubMed papers about Alzheimer’s disease and gut microbiome. This paper extends the approach by proposing a methodology, implemented as a repeatable workflow in a web-based text mining platform TextFlows, which enables easy access and execution of the methodology for the interested researcher.

A major challenge for next generation data mining systems is creative knowledge discovery from diverse and distributed data sources. In this task an important challenge is information fusion of diverse mainly unstructured representations into a unique knowledge format. This chapter focuses on merging information available in text documents into an information network --- a graph representation of knowledge. The problem addressed is how to efficiently and effectively produce an information network from large text corpora from at least two diverse, seemingly unrelated, domains. The goal is to produce a network that has the highest potential for providing yet unexplored cross domain links which could lead to new scientific discoveries. The focus of this work is better identification of important domain bridging concepts that are promoted as core nodes around which the rest of the network is formed. The evaluation is performed by repeating a discovery made on medical articles in the migraine magnesium domain.

In bisociative cross-domain literature mining the goal is to identify interesting terms or concepts which relate different domains. This chapter reveals that a majority of these domain bridging concepts can be found in outlier documents which are not in the mainstream domain literature. We have detected outlier documents by combining three classification-based outlier detection methods and explored the power of these outlier documents in terms of their potential for supporting the bridging concept discovery process. The experimental evaluation was performed on the classical migraine-magnesium and the recently explored autism-calcineurin domain pairs.

In literature mining, the identification of bridging concepts that link two diverse domains has been shown to be a promising approach for finding bisociations as distinct, yet unexplored cross-domain connections which could lead to new scientific discoveries. This chapter introduces the system CrossBee (on line Cross-Context Bisociation Explorer) which implements a methodology that supports the search for hidden links connecting two different domains. The methodology is based on an ensemble of specially tailored text mining heuristics which assign the candidate bridging concepts a bisociation score. Using this score, the user of the system can primarily explore only the most promising concepts with high bisociation scores. Besides improved bridging concept identification and ranking, CrossBee also provides various content presentations which further speed up the process of bisociation hypotheses examination. These presentations include side-by-side document inspection, emphasizing of interesting text fragments, and uncovering similar documents. The methodology is evaluated on two problems: the standard migraine-magnesium problem well-known in literature mining, and a more recent autism-calcineurin literature mining problem.

In the field of autism, an enormous increase in available information makes it very difficult to connect fragments of knowledge into a more coherent picture. We present a literature mining method, RaJoLink, to search for matched themes in unrelated literature that may contribute to a better understanding of complex pathological conditions, such as autism. 214 full text articles on autism, published in PubMed, served as a source of data. Using ontology construction, we identified the main concepts of what is already known about autism. Then, the RaJoLink method, based on Swanson's ABC model, was used to reveal potentially interesting, but not yet investigated, connections between different concepts in research. Among the more interesting concepts identified with RaJoLink in our study were calcineurin and NF-kappaB. Both terms can be linked to neuro-immune abnormalities in the brain of patients with autism. Further research is needed to provide stronger evidence about calcineurin and NF-kappaB involvement in autism. However, the analysis presented confirms that this method could support experts on their way towards discovering hidden relationships and towards a better understanding of the disorder.

A public health-care system (HCS) is a complex system that requires permanent monitoring. This paper focuses on the Slovenian national HCS sub-system consisting of a network of health-care professionals at the primary care level. The challenge addressed in this paper is the development and application of intelligent data analysis, decision support and visualization methods aimed to improve the monitoring of human resources of this network. The main outcome is a set of proposed performance indicators and the developed model for monitoring the network of primary health-care professionals of Slovenia. The model enables improved planning and management through data analysis and visualization modules developed for the monitoring of physicians' qualification, age, workload and dispersion.

This paper proposes an innovative use of data mining and visualization techniques for decision support in planning and regional-level management of Slovenian public health-care. Data mining and statistical techniques were used to analyze databases collected by a regional Public Heath Institute. We also studied organizational aspects of public health resources in the selected Celje region with the objective to identify the areas that are atypical in terms of availability and accessibility of public health services for the population. The most important step was the detection of outliers and the analysis of availability and accessibility deviations. The results are applicable to health-care planning and support in decision making by local and regional health-care authorities. In addition to the practical results, which are directly useful for decision making in planning of the regional health-care system, the main methodological contribution of the paper are the developed visualization methods that can be used to facilitate knowledge management and decision making processes.

In this article we focus on interoperable geographic information (GI) services from the crisis management perspective. Based on Open Geospatial Consortium [10] standards and initiatives, we present the building blocks of the interoperable solution for supporting crisis management that is proposed as a result of the EU sponsored project, MEDSI [1]. In particular, we focus on the application and operationalization of several OGC standards, some adopted and some still under discussion, such as WMS, WFS, WMC, SLD and SMS, as well as their integration and cooperation within a single software framework.

A system for supporting management decisions in the allocation of housing loans is presented. The system has been used in the Housing Fund of the Republic of Slovenia since 1991 for granting loans to citizens. Various activities are supported, such as priority ranking of applications, financial evaluation and analyses. The system is based on a knowledge base that contains several qualitative and quantitative decision models, and is combined with a database. In the paper, we present the design of the system, the principal stages of its development and utilization, and practical experience obtained in 11 completed floats of loans.

Rule induction, a subarea of machine learning, is concerned with the problem of constructing rules from examples. In rule induction systems, various heuristic functions are used to estimate the quality of rules. Most of them use some form of probability estimates, relative frequency being the most common. This has resulted in the problem of small disjuncts, where specific rules produce high error rates, due to unreliable probability estimates from small samples. Tu alleviate this problem, the Laplace estimate has been used in the rule induction system CN2. We have replaced the Laplace estimate by a general Bayesian probability estimate, the m-estimate, which does not rely on the Laplacian assumption of equally likely classes. The parameter m in the m-estimate allows for adapting to the learning domain. Depending on the level of noise in the examples and other properties of the domain, the appropriate level of generalization can be achieved by setting the m parameter to an appropriate value. We compare the performance of rules derived by using the Laplace and the m-estimate on several practical domains in terms of classification accuracy and the theoretically underpinned measure of relative information score.

Background In biology and medicine, experts are challenged daily with linking information from various highly specialized subfields. The individual subfields can be considered habitually different domains since experts usually master only one of them. However, many novel discoveries are achieved by gaining new insights and knowledge via fusing two or more diverse fields. In this work we propose a method that reveals key concepts which are the most informative and promising to pursue when bridging