Conversational agents offer possibilities to effectively support scenarios like health coaching to promote awareness and behavioral change. However, the process of designing conversational agents requires technical knowledge, posing challenges in onboarding domain experts such as health coaches and counselors, healthcare educators, etc. We propose a metamodeling-based domain-specific language for designing health coaching agents and conversations. Our modeling method provides elements specialized for health coaching while remaining adaptable for other coaching domains. We demonstrate the modeling language through the use case of an HIV health coaching agent. Through a feature comparison with existing model-driven approaches for chatbot development, we highlight the technology-neutral and application domain-focused contribution of our visual modeling method. Initial feedback from HIV domain experts substantiates its feasibility and relevance.

Navigating the research process, from problem identification to argumentation construction, challenges novice researchers. This study introduces RPML (Research Problem Modeling Language), a metamodel and ontology designed to address these challenges by visually representing key aspects of research argumentation. RPML enhances clarity and coherence in research discourse by providing researchers with a visual representation of argumentation for a research problem. RPML is represented as a specialization of the OMG Business Motivation Model and Toulmin's argumentation model approach. This enables researchers to gain a comprehensive overview of their research projects, identify research problems, build robust argumentation, and select suitable research strategies.

This study delves into the application of large language models (LLMs), such as ChatGPT-4, for the automated evaluation of student essays, with a focus on a case study conducted at the Swiss Institute of Business Administration. It explores the effectiveness of LLMs in assessing German-language student transfer assignments, and contrasts their performance with traditional evaluations by human lecturers. The primary findings highlight the challenges faced by LLMs in terms of accurately grading complex texts according to predefined categories and providing detailed feedback. This research illuminates the gap between the capabilities of LLMs and the nuanced requirements of student essay evaluation. The conclusion emphasizes the necessity for ongoing research and development in the area of LLM technology to improve the accuracy, reliability, and consistency of automated essay assessments in educational contexts.

Identification of a research-worthy problem, choosing the appropriate research design, and writing based on the conducted research high-quality research paper with a well-structured argumentation is a complex and multifaceted process. Therefore, employing a comprehensive approach when identifying a research problem and composing research papers is crucial to structuring the process. Design Science Research (DSR) is a commonly applied strategy in research in Information Systems. Even though this methodology has a detailed description of every research phase, for some novice researchers it is challenging to have a comprehensive overview of their research project, from the argumentation of the studied problem to the evaluation of the results. We claim that conceptual modeling can visually represent these argumentations and key elements of DSR. We provide a metamodel to represent relevant aspects of the research and the connections between them, which might be depicted as graphical elements and connectors. This approach significantly improves understanding and structuring of the research problem and design, leveraging on literature about argumentation and motivation modeling.

In Enterprise Architecture Management (EAM), rules, constraints, and principles guide and govern the Enterprise Architecture (EA). These can be formulated and verified in ontology-based enterprise architecture models. The automatic validation of EA principles relies on the knowledge available in the EA models. However, there is knowledge implicit in models that humans may understand but machines cannot. For example, relationships between model elements may be derived using derivation rules and domain knowledge. Formalizing derivation rules in an enterprise ontology, we can infer this implicit knowledge and make it available to the machine for reasoning. This research demonstrates the feasibility of using derivation rules to extract implicit knowledge from enterprise models allowing EA principles validation and supporting EAM. The research contribution is presented using a concrete real-world use case and implementing the derivation rules for the EA modeling standard ArchiMate.

Process mining is a technique used to analyze and understand business processes. It uses as input the event log, a type of data used to represent the sequence of activities occurring within a business process. An event log typically contains information such as the case ID, the performed activity’s name, the activity’s timestamp, and other data associated with the activity. By analyzing event logs, organizations can gain a deeper understanding of their business processes, identify areas for improvement, and make data-driven decisions to optimize their operations. However, as the event logs contain data collected from different systems involved in the process, such as ERP, CRM, or WfMS systems, they often lack the necessary context and knowledge to analyze and fully comprehend business processes. By extending the event logs with domain knowledge, organizations can gain a more complete and accurate insight into their business processes and make more informed decisions about optimizing them. This paper presents an approach for enhancing process mining with domain knowledge preserved in domain-specific OWL ontologies. Event logs are typically stored in structured form in relational databases. This approach first converts the process data into an event log which is then mapped with ontology concepts. The ontology contains classes and individuals representing background knowledge of the domain, which supports the understanding of the data. A class for the specific activities forms the link between the event log and the ontology. In this manner, it is possible to map the domain knowledge to a particular case and activity. This allows to determine conditions that must be satisfied for executing tasks and to prune discovered process models if they are too complex. This approach is demonstrated using data from the student admission process at FHNW and has been implemented in Protégé.

Business Process Management (BPM) is a widely applied discipline in many organizations. Creating and maintaining business process models is a task that still requires much human work and is costly and cumbersome. The reuse of business process models is a solution to minimize human effort and increase quality. For reuse, appropriate process models must be discovered in a repository. Enrichment of the models with semantic annotations from domain ontologies can leverage better results for of discovering reusable process models. Although using semantically annotated business process models for the case of reuse has been mentioned and proposed in the literature, the exact requirements and implementation have yet to be analyzed in detail. This paper closes this research gap with an artifact in the form of a methodology to discover business process models. This includes a list of relevant criteria, a base ontology, possible automated annotation techniques, and a query form.

Designing courses for introductory programming courses with a heterogeneous audience (business and IT background as well) is a challenging task. In an internal project of the School of Business at the FHNW University of Applied Sciences and Arts Northwestern Switzerland (FHNW) a group of lecturers developed a concept entitled “Model-based agile development” (MBAD) which supports the learning of elementary programming concepts in an agile environment and builds the basis for advanced courses. MBAD will be used as a basic learning module for various Bachelor programs at the FHNW.

Enterprise architecture (EA) modeling gives an opportunity to have an overview of the enterprise architecture supporting business-IT alignment within the rapidly changing environment. Visual representation of enterprise architecture models is appropriate for interpretation by humans. Machines, however, cannot interpret labels associated with the model element, as well as its domain-specific concepts. To make EA models machine-interpretable, a graphical representation of models shall be connected to domain knowledge. This research demonstrates an approach to enriching the EA model of a medical institution with healthcare domain knowledge. Evaluation of the developed solution proves that a human and a machine could equally understand the ontology-based EA model.

Due to extensive spread of fake news on social and news media it became an emerging research topic now a days that gained attention. In the news media and social media the information is spread highspeed but without accuracy and hence detection mechanism should be able to predict news fast enough to tackle the dissemination of fake news. It has the potential for negative impacts on individuals and society. Therefore, detecting fake news on social media is important and also a technically challenging problem these days. We knew that Machine learning is helpful for building Artificial intelligence systems based on tacit knowledge because it can help us to solve complex problems due to real word data. On the other side we knew that Knowledge engineering is helpful for representing experts knowledge which people aware of that knowledge. Due to this we proposed that integration of Machine learning and knowledge engineering can be helpful in detection of fake news. In this paper we present what is fake news, importance of fake news, overall impact of fake news on different areas, different ways to detect fake news on social media, existing detections algorithms that can help us to overcome the issue, similar application areas and at the end we proposed combination of data driven and engineered knowledge to combat fake news. We studied and compared three different modules text classifiers, stance detection applications and fact checking existing techniques that can help to detect fake news. Furthermore, we investigated the impact of fake news on society. Experimental evaluation of publically available datasets and our proposed fake news detection combination can serve better in detection of fake news.

This research demonstrates the feasibility of using enterprise ontologies as representation formalism for automated validation of Enterprise Architecture (EA) principles. EA Principles are usually described in natural language, while the Enterprise Architecture is represented with graphical models. As both are intended for human interpretation, it requires humans to validate whether EA Principles are satisfied by an Enterprise Architecture. This is a complex and time-consuming process. We describe a step-by-step procedure for transforming the knowledge of Enterprise Architecture models and the vocabulary of EA Principles into the ArchiMEO ontology, which allows for automated validation. The approach is evaluated in two case studies. They confirm that an ontology-based validation of EA Principles is applicable in reality.

This work aims to investigate if social media data, Twitter in particular can be used to detect early warning indicators of COVID-19 pandemic in the United States (US). To demonstrate the viability of this work, English tweets were collected with a hasghtag of COVID-19 related topics ranges from 12th March to end of April 2020. With the help of with N-gram language model and Term Frequency and Inverse Document Frequency (TF-IDF) significant bi-grams such as (“new york”), (“social, distancing”), (“stay, safe”), (“toilet, paper”), (“wash, hand”), (“tested, positive”), (look, like), (“front, line”), (“grocery, store”) etc. are extracted. Our analysis shows that, the natures of the bi-grams directly reflect the characteristics of the infection cases and are almost similarly distributed over different clusters. This study also reveals that, the tweets of (“new york”) increases with (“stay, home”), (“social, distancing”), (“stay, safe”), (“look, like”) and (“tested positive”); and decreases with (“toilet, paper”). Bi-grams with such relationships are recognized as indicators and are validated with the number of infection cases on each day. Results show that, social media data can project the actual scenario of infection curve and able to detect warning indicators once the pandemic is moderately recognized.

A company's architecture can be represented by domain-specific models, which are defined by domain-specific modeling language. Since not all stakeholders are interested in the same models, dedicated views can be created to support navigation through the enterprise models. These views offer a snippet of the entire company and cover stakeholder-specific concerns. The relationships between the different views and models remain hidden and can be unveiled with much effort. The developed concept of the zoomability principle offers the ability to change the degree of detail using zoom in and out of the enterprise model. The different models and modeling languages used to express an enterprise are considered, and a form of navigation is established similar to an online map. The concept is based on two pillars, "ZoomWithin" and "Zoom into Complements". For this purpose, a metamodel was developed, which formalizes the elements used in the concept and their relationships. Developing the artifact, rules were defined that contribute to a generic approach allowing an application to another case. Furthermore, a prototype was developed, representing the zoomability principle and offering the possibility to perform zooming behavior. The artifact was evaluated through a demonstration. An additional prototype was created to demonstrate that the developed concept can be applied to a predefined set of situations.

The AAAI 2022 Spring Symposium on Machine Learning and Knowledge Engineering for Hybrid Intelligence (AAAI-MAKE 2022) brought together researchers and practitioners of the two fields to reflect on advances in combining them, and to present the first results in creating hybrid intelligence with the two AI methods. AAAI-MAKE 2022 is the fourth consecutive edition of this symposium, which combines two prominent AI approaches, symbolic and sub-symbolic AI, as hybrid AI. In such hybrid architectures, agents using different types of AI work together to solve problems where separate approaches do not provide satisfactory results, e.g., in terms of explainability and data efficiency. Explainability is needed to complement human intelligence in the AI loop, and data efficiency (learning from small data sets) is needed in many domains where data availability is limited. Hybrid approaches that combine machine learning with the use of logic can explain inferences and increase data efficiency. The combination of machine learning and knowledge engineering opens up new possibilities for the redesign of knowledge work at the interface of humans and machines, with the aim of combining complementary strengths. Knowledge workers without strong AI expertise can contribute to hybrid teams where humans and machines work synergistically to achieve common goals better in collaboration than separately. More efforts need to be made to democratize the combination of machine learning and knowledge engineering and unleash the complementary strengths. The 2022 edition was held as a hybrid event with an on-site presence at Stanford and re-mote participation. The remarkable number of submissions again showed a huge demand for combined/hybrid AI approaches that address hybrid intelligence. These proceedings are a collection of papers that contribute to the symposium’s aim of combining machine learning and knowledge engineering, hybrid intelligence / intelligent systems, as well as hybrid AI and neuro-symbolic approaches/methods.

Fake news detection research is still in the early stage as this is a relatively new phenomenon in the interest raised by society. Machine learning helps to solve complex problems and to build AI systems nowadays and especially in those cases where we have tacit knowledge or the knowledge that is not known. We used machine learning algorithms and for identification of fake news; we applied three classifiers; Passive Aggressive, Na\"ive Bayes, and Support Vector Machine. Simple classification is not completely correct in fake news detection because classification methods are not specialized for fake news. With the integration of machine learning and text-based processing, we can detect fake news and build classifiers that can classify the news data. Text classification mainly focuses on extracting various features of text and after that incorporating those features into classification. The big challenge in this area is the lack of an efficient way to differentiate between fake and non-fake due to the unavailability of corpora. We applied three different machine learning classifiers on two publicly available datasets. Experimental analysis based on the existing dataset indicates a very encouraging and improved performance.

Detection of check-worthy statements is a subtask in the fact-checking process, automation of which would decrease the time and burden required to fact-check a statement. This paper proposes an approach focused on the classification of statements into check-worthy and not check-worthy. For the current paper, a dataset is constructed by consulting different fact-checking organizations. It contains debates and speeches in the domain of politics. Thus, even the ability of check worthy approach is evaluated on this domain. It starts with extracting sentence-level and context features from the sentences, and classifying them based on these features. The feature set and context were chosen after several experiments, based on how well they differentiate check-worthy statements. The findings indicated that the context in the approach gives considerable contribution in the classification, while also using more general features to capture information from the sentences. The results were analyzed by examining all features used, assessing their contribution in classification, and how well the approach performs in speeches and debates separately to detect the check worthy statements to reduce the time and burden of fact checking process.