Witschel, Hans Friedrich
E-Mail-Adresse
Geburtsdatum
Suchergebnisse
New hybrid techniques for business recommender systems
Besides the typical applications of recommender systems in B2C scenarios such as movie or shopping platforms, there is a rising interest in transforming the human-driven advice provided, e.g., in consultancy via the use of recommender systems. We explore the special characteristics of such knowledge-based B2B services and propose a process that allows incorporating recommender systems into them. We suggest and compare several recommender techniques that allow incorporating the necessary contextual knowledge (e.g., company demographics). These techniques are evaluated in isolation on a test set of business intelligence consultancy cases. We then identify the respective strengths of the different techniques and propose a new hybridisation strategy to combine these strengths. Our results show that the hybridisation leads to substantial performance improvement over the individual methods.
Learning and engineering similarity functions for business recommenders
We study the optimisation of similarity measures in tasks where the computation of similarities is not directly visible to end users, namely clustering and case-based recommenders. In both, similarity plays a crucial role, but there are also other algorithmic components that contribute to the end result. Our suggested approach introduces a new form of interaction into these scenarios that make the use of similarities transparent to end users and thus allows to gather direct feedback about similarity from them. This happens without distracting them from their goal – rather allowing them to obtain better and more trustworthy results by excluding dissimilar items. We then propose to use the feedback in a way that incorporates machine learning for updating weights and decisions of knowledge engineers about possible additional features, based on insights derived from a summary of user feedback. The reviewed literature and our own previous empirical investigations suggest that this is the most feasible way – involving both machine and human, each in a task that they are particularly good at.
Ensuring Action: Identifying Unclear Actor Specifications in Textual Business Process Descriptions
In many organisations, business process (BP) descriptions are available in the form of written procedures, or operational manuals. These documents are expressed in informal natural language, which is inherently open to different interpretations. Hence, the content of these documents might be incorrectly interpreted by those who have to put the process into practice. It is therefore important to identify language defects in written BP descriptions, to ensure that BPs are properly carried out. Among the potential defects, one of the most relevant for BPs is the absence of clear actors in action-related sentences. Indeed, an unclear actor might lead to a missing responsibility, and, in turn, to activities that are never performed. This paper aims at identifying unclear actors in BP descriptions expressed in natural language. To this end, we define an algorithm named ABIDE, which leverages rule-based natural language processing (NLP) techniques. We evaluate the algorithm on a manually annotated data-set of 20 real-world BP descriptions (1,029 sentences). ABIDE achieves a recall of 87%, and a precision of 56%. We consider these results promising. Improvements of the algorithm are also discussed in the paper.
Practice track: a learning tracker using digital biomarkers for autistic preschoolers
Preschool children, when diagnosed with Autism Spectrum Disorder (ASD), often ex- perience a long and painful journey on their way to self-advocacy. Access to standard of care is poor, with long waiting times and the feeling of stigmatization in many social set- tings. Early interventions in ASD have been found to deliver promising results, but have a high cost for all stakeholders. Some recent studies have suggested that digital biomarkers (e.g., eye gaze), tracked using affordable wearable devices such as smartphones or tablets, could play a role in identifying children with special needs. In this paper, we discuss the possibility of supporting neurodiverse children with technologies based on digital biomark- ers which can help to a) monitor the performance of children diagnosed with ASD and b) predict those who would benefit most from early interventions. We describe an ongoing feasibility study that uses the “DREAM dataset”, stemming from a clinical study with 61 pre-school children diagnosed with ASD, to identify digital biomarkers informative for the child’s progression on tasks such as imitation of gestures. We describe our vision of a tool that will use these prediction models and that ASD pre-schoolers could use to train certain social skills at home. Our discussion includes the settings in which this usage could be embedded.
Workplace Learning - Providing Recommendations of Experts and Learning Resources in a Context-sensitive and Personalized Manner
Support of workplace learning is increasingly important as change in every form determines today's working world in industry and public administrations alike. Adapt quickly to a new job, a new task or a new team is a major challenge that must be dealt with ever faster. Workplace learning differs significantly from school learning as it should be strictly aligned to business goals. In our approach we support workplace learning by providing recommendations of experts and learning resources in a context-sensitive and personalized manner. We utilize user s' workplace environment, we consider their learning preferences and zone of proximal development, and compare required and acquired competencies in order to issue the best suited recommendations. Our approach is part of the European funded project Learn PAd. Applied research method is Design Science Research. Evaluation is done in an iterative process. The recommender system introduced here is evaluated theoretically based on user requirements and practically in an early evaluation process conducted by the Learn PAd application partner.
Hybrid conversational AI for intelligent tutoring systems
We present an approach to improve individual and self-regulated learning in group assignments. We focus on supporting individual reflection by providing feedback through a conversational system. Our approach leverages machine learning techniques to recognize concepts in student utterances and combines them with knowledge representation to infer the student’s understanding of an assignment’s cognitive requirements. The conversational agent conducts end-to-end conversations with the students and prompts them to reflect and improve their understanding of an assignment. The conversational agent not only triggers reflection but also encourages explanations for partial solutions.
KPIs 4 Workplace Learning
Enterprises and Public Administrations alike need to ensure that newly hired employees are able to learn the ropes fast. Employers also need to support continuous workplace learning. Work-place learning should be strongly related to business goals and thus, learning goals should direct-ly add to business goals. To measure achievement of both learning and business goals we pro-pose augmented Key Performance Indicators (KPI). In our research we applied model driven engineering. Hence we developed a model for a Learning Scorecard comprising of business and learning goals and their KPIs represented in an ontology. KPI performance values and scores are calculated with formal rules based on the SPARQL Inferencing Notation. Results are presented in a dashboard on an individual level as well as on a team/group level. Requirements, goals and KPIs as well as performance measurement were defined in close co-operation with Marche Region, business partner in Learn PAd.