Refine
Departments, institutes and facilities
Document Type
- Conference Object (24)
- Article (8)
- Report (4)
- Part of a Book (1)
- Preprint (1)
Year of publication
Has Fulltext
- no (38)
Keywords
- Katamnese (3)
- Benutzeroberfläche (2)
- Concurrent Kleene Algebra (2)
- Laws of programming (2)
- Outcomeforschung (2)
- Patient-Reported Outcomes (PRO) (2)
- Refinement (2)
- Semantic models (2)
- Sozialversicherungsbeiträge (2)
- Trace algebra (2)
Tracelets and Specifications
(2017)
In the accompanying paper [1] the authors study a model of concurrent programs in terms of events and a dependence relation, i.e., a set of arrows, between them. There also two simplifying interface models are presented; they abstract in different ways from the intricate network of internal points and arrows of program components. This report supplements [1] by presenting full proofs for the properties of the interface models, in particular, that both models exhibit homomorphic behaviour w.r.t. sequential and concurrent composition. [1] B. Möller, C.A.R. Hoare, M.E. Müller, G. Struth: A discrete geometric model of concurrent program execution. In H. Zhu, J. Bowen: Proc. UTP 16. LNCS 10134. Springer 2017, 1-25
A trace of the execution of a concurrent object-oriented program can be displayed in two-dimensions as a diagram of a non-metric finite geometry. The actions of a programs are represented by points, its objects and threads by vertical lines, its transactions by horizontal lines, its communications and resource sharing by sloping arrows, and its partial traces by rectangular figures.
In this paper, we describe an approach that enables an autonomous system to infer the semantics of a command (i.e. a symbol sequence representing an action) in terms of the relations between changes in the observations and the action instances. We present a method of how to induce a theory (i.e. a semantic description) of the meaning of a command in terms of a minimal set of background knowledge. The only thing we have is a sequence of observations from which we extract what kinds of effects were caused by performing the command. This way, we yield a description of the semantics of the action and, hence, a definition.
The problem of filtering relevant information from the huge amount of available data is tackled by using models of the user's interest in order to discriminate interesting information from un-interesting data. As a consequence, Machine Learning for User Modeling (ML4UM) has become a key technique in recent adaptive systems. This article presents the novel approach of conceptual user models which are easy to understand and which allow for the system to explain its actions to the user. We show that ILP can be applied for the task of inducing user models from even sparse feedback by mutual sample enlargement. Results are evaluated independently of domain knowledge within a clear machine learning problem definition. The whole concept presented is realized in a meta web search engine, OySTER.
Machine learning seems to offer the solution to many problems in user modelling. However, one tends to run into similar problems each time one tries to apply out-of-the-box solutions to machine learning. This article closely relates the user modelling problem to the machine learning problem. It explicates some inherent dilemmas that are likely to be overlooked when applying machine learning algorithms in user modelling. Some examples illustrate how specific approaches deliver satisfying results and discuss underlying assumptions on the domain or how learned hypotheses relate to the requirements on the user model. Finally, some new or underestimated approaches offering promising perspectives in combined systems are discussed. The article concludes with a tentative "checklist" that one might like to consider when planning to apply machine learning to user modelling techniques.
The World Wide Web (Www) offers a huge number of documents which deal with information concerning nearly any topic. Thus, search engines and meta search engines currently are the key to finding information. Search engines with crawler based indexes vary in recall and offer a very bad precision. Meta search engines try to overcome these lacks by simple methods for information extraction, information filtering and integration of heterogenous information resources. Only few search engines employ intelligent techniques in order to increase precision.
Machine Learning seems to offer the solution to the central problem in recommender systems: Learning to recommend interesting items from observations. However, one tends to run into similar problems each time one tries to apply out-of-the-box solutions from Machine Learning. This article relates the problem of recommendation by user modeling closely to the machine learning problem and explicates some inherent dilemmas. A few examples will illustrate specific approaches and discuss underlying assumptions on the domain or how learned hypotheses relate to requirements on the user model. The article concludes with a tentative 'checklist' that one might like to consider when thinking about to use Machine Learning in User Adaptive environments such as recommender systems.
Roughness by Residuals
(2015)
Rough set theory (RST) focuses on forming posets of equivalence relations to describe sets with increasing accuracy. The connection between modal logics and RST is well known and has been extensively studied in their relation algebraic (RA) formalisation. RST has also been interpreted as a variant of intuitionistic or multi-valued logics and has even been studied in the context of logic programming.
On nothing
(2014)
Learning Adaptive Behavior
(2005)
This work presents the preliminary research towards developing an adaptive tool for fault detection and diagnosis of distributed robotic systems, using explainable machine learning methods. Autonomous robots are complex systems that require high reliability in order to operate in different environments. Even more so, when considering distributed robotic systems, the task of fault detection and diagnosis becomes exponentially difficult.
To diagnose systems, models representing the behaviour under investigation need to be developed, and with distributed robotic systems generating large amount of data, machine learning becomes an attractive method of modelling especially because of its high performance. However, with current day methods such as artificial neural networks (ANNs), the issue of explainability arises where learnt models lack the ability to give explainable reasons behind their decisions.
This paper presents current trends in methods for data collection from distributed systems, inductive logic programming (ILP); an explainable machine learning method, and fault detection and diagnosis.
Ein wichtiges Ziel der medizinischen Rehabilitation der Rentenversicherung war schon immer die berufliche (Wieder-)Eingliederung der Rehabilitanden. Daher ist die Ermittlung des Erwerbsstatus ein zentrales Element für die Bewertung des Rehabilitationsergebnisses. Für die konkrete Umsetzung bestehen jedoch mehrere Möglichkeiten: Betrachtung von Einkommen, Beschäftigungsdauer oder -art, Stichtags- oder Zeitraumbetrachtung, kurz-, mittel- oder langfristige Erhebung, Befragung der Rehabilitanden oder Nutzung von Daten der Sozialversicherung etc. In diesem Beitrag werden mögliche Herangehensweisen am Beispiel der „Reha-QM-Outcome-Studie“ des „Qualitätsverbundes Gesundheit“ und der Deutschen Rentenversicherung Baden-Württemberg (Kaluscha et al., 2014) beleuchtet.
In der Ergebnisdarstellung der Reha-QM-Outcome Studie der DRV Baden-Württemberg und des Qualitätsverbunds Gesundheit konnte gezeigt werden, dass Kliniken eines Verbundes, die ein gemeinsames, auf aktivem Benchmarking und Von-Einander-Lernen gegründetes Qualitätsmanagement (Basis QMS Reha®) anwenden, ein Jahr nach der Reha etwas bessere Ergebnisse in relevanten Outcome-Parametern (u. a. subjektiver Reha-Nutzen, geleistete Rentenversicherungsbeiträge) erzielen als der Durchschnitt der Kliniken (Toepler et al., 2015). Der vorliegende Beitrag stellt die verbundinterne Analyse der Studienergebnisse dar und geht der Frage nach, welche QM-Elemente einen positiven Einfluss auf die Outcome-Parameter ausüben.
Return to Work (RTW) stellt ein wesentliches Outcomekriterium für die Abbildung der Effektivität medizinischer Rehabilitationsmaßnahmen dar. Dabei hängt die Höhe der RTW-Quote u. a. von Messmethode, Messzeitpunkt und Stichprobe ab (Streibelt, Egner, 2012). RTW wird häufig mit dem bloßen Status der Erwerbstätigkeit oder Arbeitsfähigkeit gleichgesetzt, wobei kritisiert werden kann, dass dabei der Aspekt einer dauerhaften beruflichen Wiedereingliederung zu wenig Berücksichtigung findet.
Neben der Verbesserung des Gesundheitszustandes sind der Erhalt der Beschäftigungsfähigkeit und die berufliche (Wieder-)Eingliederung zentrale Ziele der Rehabilitationsleistungen der Deutschen Rentenversicherung. In der „Reha-QM-Outcome-Studie“ wurden sowohl mittels Patientenfragebogen Angaben zum subjektiven Nutzen der Behandlung als auch mittels Routinedaten der Rentenversicherung Angaben zum Erwerbsstatus erhoben, so dass eine Gegenüberstellung beider Zieldimensionen erfolgen kann.
Internes Qualitätsmanagement (QM) wurde spätestens 2007 mit dem Gesetz zur Stärkung des Wettbewerbs in der gesetzlichen Krankenversicherung zu einem wesentlichen Bestandteil der stationären medizinischen Rehabilitation (Petri, Stähler, 2008). Seit dem Auslaufen der Übergangsfrist am 01.10.2012 verfügen alle durch einen gesetzlichen Rehabilitationsträger belegten stationären Einrichtungen über ein, den Anforderungen der Bundesarbeitsgemeinschaft für Rehabilitation entsprechendes, zertifiziertes QM-System.
Die Ergebnisqualität medizinischer Rehabilitationsleistungen wird häufig über „Patient Reported Outcomes“ (PROs) gemessen. Die Bedeutung von PROs für die Nutzenbeurteilung von therapeutischen Interventionen wird häufig unterschätzt (Brettschneider et al., 2011; Calvert et al., 2013). Es wird untersucht, inwieweit sich PROs in „harten“ Endpunkten wie z. B. Beitragszahlungen der Versicherten in die Sozialversicherung widerspiegeln.