While learning personalization offers great potential for learners, modern practices in higher education require a deeper consideration of domain models and learning contexts, to develop effective personalization algorithms. This paper introduces an innovative approach to higher education curriculum modelling that utilizes large language models (LLMs) for knowledge graph (KG) completion, with the goal of creating personalized learning-path recommendations. Our research focuses on modelling university subjects and linking their topics to corresponding domain models, enabling the integration of learning modules from different faculties and institutions in the student's learning path. Central to our approach is a collaborative process, where LLMs assist human experts in extracting high-quality, fine-grained topics from lecture materials. We develop a domain, curriculum, and user models for university modules and stakeholders. We implement this model to create the KG from two study modules: Embedded Systems and Development of Embedded Systems Using FPGA. The resulting KG structures the curriculum and links it to the domain models. We evaluate our approach through qualitative expert feedback and quantitative graph quality metrics. Domain experts validated the relevance and accuracy of the model, while the graph quality metrics measured the structural properties of our KG. Our results show that the LLM-assisted graph completion approach enhances the ability to connect related courses across disciplines to personalize the learning experience. Expert feedback also showed high acceptance of the proposed collaborative approach for concept extraction and classification.

In the era of personalized education, the provision of comprehensible explanations for learning recommendations is of great value to enhance the learner's understanding and engagement with the recommended learning content. Large language models (LLMs) and generative AI have recently opened new doors for generating human-like explanations, for and along learning recommendations. However, their precision is still far away from acceptable in a sensitive field like education. To harness the abilities of LLMs, while still ensuring a high level of precision towards the intent of the learners, this paper proposes an approach to utilize knowledge graphs (KG) as a source of factual context for LLM prompts, reducing the risk of model hallucinations, and safeguarding against wrong or imprecise information, while maintaining an application-intended learning context. We utilize the semantic relations in the knowledge graph to offer curated knowledge about learning recommendations. With domain-experts in the loop, we design the explanation as a textual template, which is filled and completed by the LLM. Domain experts were integrated in the prompt engineering phase as part of a study, to ensure that explanations include information that is relevant to the learner. We evaluate our approach quantitatively using Rouge-N and Rouge-L measures, as well as qualitatively with experts and learners. Our results show an enhanced recall and precision of the generated explanations compared to those generated solely by the GPT model, with a greatly reduced risk of generating imprecise information in the final learning explanation.

In emergencies, treatment needs to be fast, accu-rate and patient-specific. For instance, in emergency scenarios, obstacles like treatment environments and medical difficulties can lead to bad outcomes for patients. Additionally, a drastic change of health vitals can force paramedics to shift to a different treatment in the ongoing treatment of the patient in order to save a patient's life. The KIRETT (engl.: 'Artificial intelligence in rescue operations ‘) demonstrator is developed to provide a rescue operator with a wrist-worn device, enabling treatment recommendation (with the help of knowledge graph) with situation detection models to improve the emergency treatment of a patient. This paper aims to provide a qualitative evaluation of the 2-days testing in the KIRETT project with the focus of knowledge graphs, knowledge fusion, and user-experience-design (UX-design).

Modelling learning objects (LO) within their context enables the learner to advance from a basic, remembering-level, learning objective to a higher-order one, i.e., a level with an application- and analysis objective. While hierarchical data models are commonly used in digital learning platforms, using graph-based models enables representing the context of LOs in those platforms. This leads to a foundation for personalized recommendations of learning paths. In this paper, the transformation of hierarchical data models into knowledge graph (KG) models of LOs using text mining is introduced and evaluated. We utilize custom text mining pipelines to mine semantic relations between elements of an expert-curated hierarchical model. We evaluate the KG structure and relation extraction using graph quality-control metrics and the comparison of algorithmic semantic-similarities to expert-defined ones. The results show that the relations in the KG are semantically comparable to those defined by domain experts, and that the proposed KG improves representing and linking the contexts of LOs through increasing graph communities and betweenness centrality.

This paper provides a brief overview about the latest applications in blockchain domain especially the trends and research questions giving an idea about limitations and benefits and in conclusion a perspective for future work. The goal is to develop open topics for research or business usage and discuss a basic idea of a possible future blockchain approach that can be utilized for the process of interchanging knowledge.

Artificial intelligence (AI) and intelligent personal assistants (IPAs) are becoming more and more important. This is no longer limited to private use but also becoming increasingly important in everyday business life. The identification of optimization potentials through the use of AI and IPAs is therefore relevant from both a theoretical and a practical point of view. This paper, therefore, identifies concrete use cases for relevant processes in small- and medium sized enterprises (SME) in the service industry and enhances them with AI and IPA capabilities. Based on a prototype, the use cases were presented to 10 experts who were interviewed regarding their usefulness and influencing factors. Subsequently, the results of the interviews were categorized and validated again by a quantitative survey within the expert panel. As a result, the use cases were evaluated with regard to the specific influencing factors and the potential for optimization was determined. The use cases were evaluated based on this data. It was shown that IPA features in particular are perceived as useful. On average, AI and IPA features have a cost savings potential of over 31%. This shows the importance of these features and the need to consider them when modeling modern business processes.

Bug reports are essential software artifacts that describe software bugs using natural language. Bug localization tools can help developers to understand the relation between bug reports and a software system. However, most approaches for localizing bugs work with unstructured textual information from the source codes and bug reports. This paper proposes an approach for locating and visualizing bug reports based on class diagrams representing the overall structural design of a software system. Our approach called IdentiBug takes advantage of deep learning techniques to train our bug localization model to predict connections between a bug report and the system’s class diagram. The result is a ranked list of classes from which we extract and rank a list of class diagram excerpts for assisting the developers during bug documentation and localization.

The growth of online job-posting repositories provided job-seekers with access to a large number of potential jobs. User assessment of recommended jobs becomes especially a tedious and time-consuming task with the overwhelming number of job recommendations. To enhance the job-seeker’s ability to evaluate the suitability of a recommended job, we propose an explainable job recommendation system, which matches the user to the most relevant jobs based on their profile. Then, the system explains to the user why each job-posting has been recommended to them. The proposed system uses a knowledge graph (KG) structure to model job-postings and user profiles in one homogeneous structure. Graph relations between the job-seekers and job-postings are mined through natural language processing (NLP) of the textual content from job-postings and user-profiles. Based on the graph structure itself and a customized named entity classifier, a human-readable explanation is generated for each recommendation and provided to the job-seeker. The explanation includes information about the matching factors that led the system to recommend a certain job-posting to the user. The proposed system is implemented and tested on a sample data-set of user profiles and job-postings from open online repositories. We use BELU and Rouge-L scores to show that the proposed systems generated relevant explanations for recommended jobs.

In recent years, the availability of modern technology increased drastically with the raising availability of integrated devices and applications, such as mobile phones, voice assistant systems smart home appliances and many more. This paved the way for the semiconductor industry to become one of the fastest growing industries. Knowing, planning, and stabilizing the yield of semiconductor manufacturing is highly important to meet the rising demand. One indication for potential root causes is the identification of defect patterns on wafers. Wafers are the base silicon layer on which sets of chips are manufactured. If a certain process damages chips, then this produces characteristic patterns of failing chips on the wafer, which are then investigated by engineers to isolate the root-cause. According to studies, human-expert based defect pattern recognition methods have a maximum accuracy of about 45%. To help engineers to improve recognition and root cause analysis of defect patterns, this paper introduces a novel process for analysis. For this, unsupervised machine learning and clustering techniques are utilized to identify and group unknown defect patterns. A tailored process is introduced, using autoencoders and an iterative classification, which is tested on a use case with a pre-known root cause.

BACKGROUND: Melanoma results in the vast majority of skin cancer deaths during the last decades, even though this disease accounts for only one percent of all skin cancers' instances. The survival rates of melanoma from early to terminal stages is more than fifty percent. Therefore, having the right information at the right time by early detection with monitoring skin lesions to find potential problems is essential to surviving this type of cancer. RESULTS: An approach to classify skin lesions using deep learning for early detection of melanoma in a case-based reasoning (CBR) system is proposed. This approach has been employed for retrieving new input images from the case base of the proposed system DePicT Melanoma Deep-CLASS to support users with more accurate recommendations relevant to their requested problem (e.g., image of affected area). The efficiency of our system has been verified by utilizing the ISIC Archive dataset in analysis of skin lesion classification as a benign and malignant melanoma. The kernel of DePicT Melanoma Deep-CLASS is built upon a convolutional neural network (CNN) composed of sixteen layers (excluding input and ouput layers), which can be recursively trained and learned. Our approach depicts an improved performance and accuracy in testing on the ISIC Archive dataset. CONCLUSIONS: Our methodology derived from a deep CNN, generates case representations for our case base to use in the retrieval process. Integration of this approach to DePicT Melanoma CLASS, significantly improving the efficiency of its image classification and the quality of the recommendation part of the system. The proposed method has been tested and validated on 1796 dermoscopy images. Analyzed results indicate that it is efficient on malignancy detection.

The maintenance and repair of modern vehicles is a challenge for garages, as different causes of faults lead to similar symptoms in the highly complex vehicles these days. Existing processes for fault-diagnosis based on manufacturer service manuals and human experiences are often inadequate and result in high effort and wrong decisions. In addition to these service manuals which provide basic models for e.g., diagnostic terms, primary physical quantities, causal relationships, and plausibilities, nowadays, internet forums offer a comprehensive source of experiences for solutions to these challenges. This paper, therefore, presents methods for the extraction of knowledge from unstructured and informal contributions in internet forums with the goal to synthesize diagnostic graphs from the established knowledge base, which are part of a maintenance software to supports garages in the maintenance of vehicles by suggesting more efficient and target-oriented diagnostic and maintenance actions in real-time.

The visualization and simplification of complex semantically-related knowledge is one of the main challenges in knowledge discovery. In this regard, the knowledge map is a good visualization instrument to represent and provide suitable information with analysis potential. Multidimensional knowledge bases aim to support this objective and store automatically extracted facts and their dimensional relations from textual knowledge resources. In this paper, a dynamic layout structure for knowledge maps based on dimensional information is introduced. The Concept of the Imitation of the Mental Ability of Word Association (CIMAWA) is applied in this approach to create a graphical structure as arrangement of associated information on different levels of textual information.

Current package inserts for medicines are confusing for many patients and can lead to non-compliance. To combat this problem, a readability assistance system has been designed to analyze and improve leaflets to reduce the risks for patients, such as incorrect use of medication. This assistance system is divided into 5 levels: pharmaceutical readability index, graphical package insert leaflet overview, paragraph analysis, sentence analysis and recommendation of sentence rearrangement. The assistance system is designed only as a guide for the person in charge of the Patient Information Leaflet (PIL) to guard against possible misunderstanding of the PIL. The final decision is always with the person in charge for the correctness of the wording. In the future it is planned to adapt the assistance system to other types of leaflets.

As financial markets getting faster and more complex, it is difficult for market participants to manage the information overload. Sentiment analysis is a useful text mining method to process textual content and filter the results with analysis methods to relevant and meaningful information. The paper in hand introduces a new method for sentiment analysis in financial markets which combines word associations and lexical resources. Based on stock market news from January 2000 to February 2014 we analyzed documents on different levels. The results are presented and evaluated in this paper.

Managers of enterprises concern with a major challenge for optimal management of human resources based on availability of domain experts and highly qualified personnel. The process of allocating right people to the right positions in a right time is a key to success. To achieve this goal, managers need to deploy evaluation tools integrated with the gap analysis method. This paper presents the concept and implementation details of an in-house developed software tool for competence evaluation of domain specific competencies and selection of professionals. A generic mathematical representation of competences in this project makes the software tool applied in a wide variety of organizations. A standard competence model has been first defined in this project with 5 main competence categories and related sub-categories including over 70 competence questionnaires in different managerial and employee levels. Test and evaluation of the software have been carried out by initializing the lab data of over 50 candidates with student groups involved in the project at the institute of Knowledge Based Systems and Knowledge Management, University of Siegen. The paper reflects the conception and the outcomes of the implementation of the software tool. The ultimate objective of this interdisciplinary project is to fill the gap in the selection process by means of an efficient and practical competency evaluation tool. The generic software tool is aimed to be used as a component in research and industrial projects of the institute.

This paper discusses design, development and prototyping of a drive designer assistance tool: MainDriveASSIST. The prototype consists of features that enable drive engineer in selection of drive system components based on meta-analysis of vendors' offers as well as customer's requirements. Meta-analysis is performed through by using benchmarking dashboard comprises technical and non-technical indicators for selection of drive system components. The benchmarking dashboard broadens and deepens drive engineer's knowledge for optimal selection of drive components and improves decision-making activities. Technical indicators are mainly acquired from dimensioning of the drive system (drive system specifications) as well as the basic requirements e.g. for power consumption and operational availability (maintenance related factors). Non-technical indicators are mainly to consider business administration activities especially through indication of investment and preventive/corrective maintenance costs of drive systems which are directly affect customer's/company's business process. The prototype of MainDriveASSIST is designed and developed within an industrial oriented project at the institute of Knowledge Based Systems and Knowledge Management.

Enterprises and industrial companies strive to improve their functional performance by identification of core competencies in order to utilize human resources and optimise the knowledge integration processes of the company. In addition, maintenance operations are one of the most important sections in industrial companies which consist of key personnel and also explicit and implicit knowledge resources that have direct effects on product quality and return on investment. In the context of implicit knowledge resources, the principal objective is firstly to identify knowledge holders who are mainly domain experts (e.g. Chief Maintenance Officer-CMO) and maintenance practitioners (e.g. engineers, technicians, etc.), and secondly to measure their domain expertise. This paper presents the concept and implementation results of an interdisciplinary research which aims at improving the knowledge measuring of maintenance practitioners. In this way, the companies are enabled to deduce rate of human failures in maintenance operations by allocating the right professionals in the right positions using competency profiling of employees. The implementation results in developing a competence profiling tool as an add-on for Computerized Maintenance Management Information Systems (CMMIS), which is previously developed in the Institute of Knowledge Based Systems and Knowledge Management (KBS&KM).

This paper introduces a comprehensive concept idea for building up a knowledge based innovation framework for scientific research projects in general and especially in the field of material science (IConMas). Regarding the aspect of material science the aim is to develop and establish a knowledge based material innovation method. Both aspects, innovation framework and innovation method, are integrated in a common concept which is accurately described within this paper. The outlook focuses on possibilities for realizing the concept and further work in the project.

This paper describes a unifying data management concept for feedback assistance systems. The assistance system is to be integrated into existing system scenery. During the use of a hydraulic system, objective feedback such as sensor data or service data is captured and managed in local databases by the machine operators. The aim is to lead such Product Use Information (PUI) back into product development. That way, the product developer uses the information to derive additional and innovative potential to improve next generation products. The information flow is related in various systems i.e. the operator of the machine and the manufacturer. Furthermore, the PUI is managed inside the assistance system based on a data warehouse. The data warehouse is coupled with a Product Lifecycle Management (PLM) system, a well-known system to product developers. Within the PLM system, product master data is created and managed. This also serves as the basis for the assistance system, so that the data models for both systems have to be amalgamated.

This paper describes projects that demonstrate applied knowledge management in medical engineering. It describes Stroke and Alzheimer portal system solutions, designed for patients and their families cope with their problems and help physicians in their treatment. A video conferencing system illustrates how specialists and general practitioners are connected superiorly. A readability test assists to prepare information in a package leaflet for patients. All described projects show that knowledge-based solutions can effectively support people in health matters. The outlook illustrates a comprehensive medical framework which facilitates knowledge transfer and network building between patients, relatives and medical experts.

Transformations into the modern business world is sustained by enhancement and improvement of strategies, systems and techniques towards evaluating and applying customer knowledge for the integration of Product Use Information (PUI) into product development, and meeting customer and market demands. In this paper the processing and modelling of PUI of many instances of one product type which is captured during the product use phase, e.g. condition monitoring data, failures or incidences of maintenance, raised by different graphical methods on the basis of a praxis and application scenario. Product Lifecycle Management (PLM) ensures a uniform data basis for supporting numerous engineering and economic organizational processes along the entire product life cycle - from the first product idea to disposal or recycling of the product. The processing and modelling of PUI raised by graphical methods like Bayesian Networks (BNs) or Dynamic Bayesian Networks (DBNs). In accordance, the product use knowledge leads back of the product development phase. This is used for discovering room for product improvements for the next product generation. Therefore the PUI of the different instances should be aggregated by applying fusion techniques to deduce/achieve generalized product improvements for a product type which is related to prospective research by focus on quality management systems and defining measures for customer satisfaction. As a result the significant aspect of this paper is to identify which graphical solution brings optimally the best results for the requirements of processing and modeling of PUI.

Improving the quality of products is an important issue in the modern business world. Traditional approaches of Quality Management (QM) are not adequate to fulfil the demands on target quality of products. This study reveals that synergetic approaches based on the integration of Knowledge Management (KM) in Total Quality Management (TQM) have a direct impact on enhancing the quality of products. We also propose a management model to synthesize elements of both methodologies under an integrative framework. Furthermore, Knowledge Discovery in Databases (KDD) is introduced to realize the effectiveness of the proposed management model and to illustrate the influence of this synergetic approach taking the semiconductor industry as exemplary field of application.

Since real product use information is not available, Design Simulation and Design Methods rely in many ways on assumptions regarding the product use today. These assumptions generally differ from the real conditions of product use. There are various reasons for not feeding back product use information. First, current business models lead to a loss of access to the product after sale. Second, due to their price and size appropriate sensors are only rarely embedded in the product. Third, there is a lack of an integrated framework for feeding back product use information into product development. This paper presents a new solution approach for the integration of product use information into product development. The first part of the paper provides a summary of the developed solution. While aspects like data management and knowledge discovery have been covered in previous work, this paper focuses on the representation of empirical product use information and the use of knowledge based inference methods in order to carry out ¿What-If¿ analyses. These can serve the product developer to improve the design of next generation products.

The key point of this paper is the proactive management of the whole risk of an IT project portfolio. A portfolio is collection of projects and every project implicates its own risk. Project risks describes how uncertainty and risk are identified, analyzed and cumulated for a single project. The most attention is paid to the quantification of risk that is based on fundamental decision theory. IT project portfolio management clarifies the sense and the benefit of a portfolio for IT projects. Especially topics like implementation of a portfolio and measurement criteria are fundamental discussion points. Portfolio management including risk as measurement tries to figure out how the whole risk for a portfolio could be made up of the single project risks. Especially inter-dependencies and correlations across the projects are considered. Also addressed is the issue of how to minimize the whole risk of an IT project portfolio by diversification. Proactive risk management and mitigation discusses all opportunities for risk mitigation, that includes reduction of the probability that a risk will materialize or reduction of the impact on the business of a risk event if it does occur.

The aim of this work is to present recent advances in porting the directed covariance matrix adaptation-evolution strategy (DCMA-ES) from the test function environment to a real-world application. DCMA-ES is a CMA-ES variant enhanced by the powerful concept of directed mutation. So far, DCMA-ES has already proven to outperform the classical CMA-ES for several test functions. Hence, the target is now to benefit from the strength of this approach also in real-world scenarios, as for example in optimizing screw-type machines. The paper is organized as follows: first we outline the DCMA-ES and the functioning of the screw-type machine. Then the performance of DCMA-ES on some test functions is presented as motivation for the porting whose current status is discussed. In the final section, conclusions are provided and the future work is sketched.

Among the various types of decision support systems, decision-theoretic models and rule-based systems have gained considerable attraction. Both approaches have advantages and disadvantages. Decision-theoretic models like decision networks dispose of a sound fundamental mathematical basis and comfortable knowledge engineering tools. Rule-based systems provide an efficient execution architecture and represent knowledge in an explicit, intelligible way. In this paper, we consider fuzzy rule-based systems as a special type of condensed decision model. We outline a knowledge transformation and compilation scheme which allows one to transform a decision-theoretic model into a fuzzy rule base and, hence, to combine the advantages of both approaches. An experimental example is given as demonstration of the described techniques.

A tool for the interactive design of screw type machines are presented that also comprises an optimization module. Beside the usual features of computer aided design tools, it has unique support for screw type machine design, as for example automated calculation of the female rotor for an arbitrary given male rotor. The optimization relies on an evolution strategy with a new mutation operator, called directed mutation. It was presented in short and some results using this optimization module are given. In the final section, conclusions are provided.

In this paper we address and discuss the problem of creating graphical models as representation format of personalized skill profiles in different application environments (e.g. skill matching for staff positioning in a project). We can learn employee skills from data based on structured profiles and their representation as Bayesian network structure using an information theoretic dependency analysis approach. Many enterprise resource management systems (ERP) come along with integrated modules for Human Resource Management (HRM). One main task of HRM is to manage, improve and deploy the right skills at the right time. These processes are well known as skill management. Furthermore the problem of finding dependencies between employee skills not obvious in evidence is considered. Using an information theoretic approach to construct a powerful skill representation as graphical model is comprehensible. To demonstrate the achievement of the learned and created network structure, a test scenario concerning historical reference project data is given.

This paper concerns the application of digital image processing in combination with rulebased reasoning as a nondestructive quality testing method for resistance spot welding. The use of fuzzy logic gives the possibility to characterize colors by linguistic terms rather than by names for wavelength or wavelength intervals. The presented method supplements Lotfi Zadeh's idea of soft computing and computing with words. It can be applied to many problems in which color carries important information.

In this paper, we present the use of stochastic learning automata (SLA) in multiagent robotics. In order to fully utilize and implement learning control algorithms in the control of multiagent robotics, an environment for simulation has to be first created. A virtual laboratory for simulation of autonomous agents, called V-Lab is described. The V-Lab architecture can incorporate various models of the environment as well as the agent being trained. A case study to demonstrate the use of SLA is presented.

The demand for high-performance computing grows with the increasing number of applications within this field. A generalized grid-computing concept may not be restricted to a single facility or a country, but may rather feature a world-wide network of distributed supercomputing clusters that are to the disposal of many research groups around the world. Each individual cluster may differ a lot from the other participating clusters in terms of computing power. Also, the computational tasks which are to be assigned to different clusters may also differ in demand of computing power. Therefore, an accurate scheduling of the diverse computational tasks to different clusters improves the cost-effective utilization of the network resources.

Evolution strategies are a powerful variant of the evolutionary algorithms, which themselves are probabilistic optimization methods. Many sophisticated methods have been developed to increase the convergence of evolution strategies. Self-adaptation is one of these methods and allows an evolution strategy to adapt to the goal function. Nevertheless most real world applications of evolution strategies do not make use of the self-adaptation. The authors analyze the reasons for this and introduce a new type of self-adaptation that overcomes the disadvantages of the known types. Experimental results based on the sphere model are presented, which show an significant increase of performance.

Fuzzy logic in the field of image processing is becoming more and more important. Image quality is improving, but data contains different kinds of uncertainty which have to be handled using computerized analysis. These uncertainties can be found particularly in medical image data. In this paper, the handling of such uncertainties using methods such as the natural language of medical experts and iconic fuzzy sets for the description and recognition of anatomical structures in magnetic resonance tomograms (MRT) of human heads are presented.

In this paper the applicability of evolution strategies, a special kind of evolutionary algorithms, to the problem of parameter optimization in the development of fuzzy rule-based systems is demonstrated. For this aim we introduce a shell which supports the design of any kind of rule based systems employing fuzzy logic for the formalization of imprecise reasoning processes and which optimizes all numerical parameters. This method works model-free, we do not need to know implicit features of the optimizing system.

In this paper concepts for goal-oriented reasoning within the blackboard development environment QBB are presented. The architecture of QBB supports the selection of problem solving actions with respect to the achivement of quality goals. Furthermore, interactions of goals are explicitly taken into account in action selection. The features of QBB to support goal-oriented reasoning are presented. Especially, it is described how mutual influence of actions with respect to goal achievement can explicitly modeled as relationships between actions, the so-called compensation relations. The usefullness of compensation relations has been tested by goal-oriented modeling of the travelling salesman problem.