Love Ekenberg

Value Differences using Second Order Distributions

Most decision models for handling vague and imprecise information are unnecessarily restrictive since they do not admit for discrimination between different beliefs in different values. This is true for classical utility theory as well as for the various interval methods that have prevailed. To allow for more refined estimates, we suggest a framework designed for evaluating decision situations considering beliefs in sets of epistemically possible utility and probability functions, as well as relations between them. The various beliefs are expressed using different kinds of belief distributions. We show that the use of such distributions allows for representation principles not requiring too hard data aggregation, but still admitting efficient evaluation of decision situations.

Agent metaphors for analysing telematic services

Using as little mathematical detail as possible, we present the use of transformations between process related languages and dynamic entity-relationship models. The work incorporates our earlier research in schema integration and the integration of multi-agent architecture designs to handle problematic cases of global inconsistency in distributed information systems. The verification methods in first order logic (FOL) presented here are generally applicable to the analysis of conflicts in process based systems. Agents are a useful metaphor for handling asynchronous processes in contexts such as telecommunications. Although the application here is one of telephonic services, the general framework when applied to agent architectures is equally suitable for extended services such as decision making agents as the basis for intelligent home services. Telia Research AB (previously Swedish Telecom) commissioned this work, the implementation of which was used to detect interferences to Telias public telematic services.

Un modelo para cursos Mega en línea. Desarrollo, implementaci ón y evaluaci ón de un curso efectivo de aprendizaje en línea a gran escala

The need for higher education on a global and local level is ever increasing due to social, political and economical factors. Universities are faced with the challenge of educating larger student bodies with the same or even reduced resources. Efficient use of ict and adapted pedagogical design are means do meet this challenge. This paper presents a large scale ict based distance course in project management given at the Department of Computer and Systems Sciences, Stockholm University, Sweden. The course has been given seven times during the period 2006-2008, with a total of 2427 students. The aim of this paper is to explore the following specific problems related to online mega courses: (1) The low student completion rate, (2) The changed pedagogical and psycho-social work situation for teachers, (3) Communication and courses suitable for this type of format, furthermore the paper gives a number of suggestions based on our experiences and obtained knowledge about mega online course. This is done through the use of data collected through the educational platform in various databases, interviews with involved teachers, and surveys filled out by students. Some of the conclusions that are presented in this paper are: Most significant changes that influenced completion rate was assignment redesign, clear sequence of events through the assignments, introduced self study courses that enables students to learn the tools needed to complete the assignments. The biggest loss of students was between registering for the course and actually starting the course and when the course moved from theoretical study to practical application. Minor design errors led to huge amount of extra work. The need for development of an effective administrator interface is crucial and must not be forgotten. The psycho-social work situation for online teachers managing mega scale courses are significantly altered as compared to class-room teaching.