Aron Larsson

Using a Software Tool for Public Decision Analysis: The Case of Nacka Municipality

This paper presents a case of interval decision analysis using a tool that takes advantage of interval probabilities, values, and criteria weights and is capable of handling comparative relations, i.e., interval statements on differences between variables. These statements are represented as constraints to the solution set and evaluated using a number of different evaluation methods, each serving the decision maker with different insights of the decision problem. We demonstrate the applicability of the tool in a case study of three public infrastructure decision problems that had remained unresolved for a number of years.

Distribution of expected utility in decision trees

Evaluation of decision trees in which imprecise information prevails is complicated. Especially when the tree has some depth, i.e. consists of more than one level, the effects of the choice of representation and evaluation procedures are significant. Second-order representation and evaluation may significantly increase a decision-makers understanding of a decision situation when handling aggregations of imprecise representations, as is the case in decision trees or influence diagrams, while the use of only first-order results gives an incomplete picture. Furthermore, due to the effects on the distribution of belief over the intervals of expected utilities, the @C-maximin decision rule seems to be unnecessarily pessimistic as the belief in neighbourhoods of points near interval boundaries is usually lower than in neighbourhoods near the centre. Due to this, a generalized expected utility is proposed. The results in this paper apply also to approaches, which do not explicitly deal with second-order information, such as standard decision trees or probabilistic networks using only first-order concepts, for example upper and lower bounds. Furthermore, the results also apply to other, non-probabilistic weighted trees such as multi-criteria weight trees.

Decision analysis with multiple objectives in a framework for evaluating imprecision

We present a decision tree evaluation method for analyzing multi-attribute decisions under risk, where information is numerically imprecise. The approach extends the use of additive and multiplicative utility functions for supporting evaluation of imprecise statements, relaxing requirements for precise estimates of decision parameters. Information is modeled in convex sets of utility and probability measures restricted by closed intervals. Evaluation is done relative to a set of rules, generalizing the concept of admissibility, computationally handled through optimization of aggregated utility functions. Pros and cons of two approaches, and tradeoffs in selecting a utility function, are discussed.

Transparent Public Decision Making - Discussion and Case Study in Sweden

The Swedish city of Orebro has since long faced complex problems with poor water quality in a local river. This problem is a typical example of a regional decision problem, since there are several different stakeholders that might be affected, and there are different views on the need for, and effect of, different measures. The problems also strongly relate to the environmental condition of the river and involve other municipalities as well. In this chapter, we describe how to address this problem using an implementation of a systematic democratic decision process for enhancing the transparency and the decision quality in itself. The process is in conformity with common democratic processes, but with higher emphasis on accuracy and precision and on the interaction between civil servants and decision makers. A main issue here is to clearly separate the various views involved in these processes from the actual facts and, at the same time, facilitate input from various stakeholders. Therefore, we allow for modelling of outcomes based on different preferences and facilitate an elicitation process where views are extracted and combined with basic data from the background investigations preceding the decision. The process is divided into two stages. The first one is emphasized in this chapter and concerns the internal democracy, i.e. the formulation and refinement of the original and extended decision problems and the interaction between politicians and civil servants, while the second stage deals with the external democracy, i.e. the communication with the public, where communication channels directed towards citizens will be formed.

A Prescriptive Approach for Eliciting Imprecise Weight Statements in an MCDA Process

In this article, we discuss decision making involving multiple objectives (MCDA) and especially the lack of more prescriptively useful elicitation methods for weights within MCDA. We highlight the discrepancy between how elicitation is handled in current decision analysis applications and the abilities of real decision-makers to provide what is required from them. Based on theory and highlighted problems with current methods, we propose a novel approach for weight elicitation which relaxes the need for numeric preciseness from decision-makers and reduces some of the practical issues related to such processes. The method is tested in a comparative study, as well as employed in a real-life case study.

Weighting Under Ambiguous Preferences and Imprecise Differences in a Cardinal Rank Ordering Process

The limited amount of good tools for supporting elicitation of preference information in multi-criteria decision analysis (MCDA) causes practical problem. In our experiences, this can be remedied by allowing more relaxed input statements from decision-makers, causing the elicitation process to be less cognitively demanding. Furthermore, it should not be too time consuming and must be able to actually use of the information the decision-maker is able to supply. In this paper, we propose a useful weight elicitation method for MAVT/MAUT decision making, which builds on the ideas of rank-order methods, but increases the precision by adding numerically imprecise cardinal information as well.

An Object-Oriented Model in Support of Context-Aware Mobile Applications

Intelligent and context-aware mobile services require users and applications to share information and utilize services from remote locations. Thus, context information from the users must be structured and be accessible to applications running in end-devices. In response to this challenge, we present a shared object-oriented meta model for a persistent agent environment. The approach enables agents to be context-aware facilitating the creation of ambient intelligence demonstrated by a sensor-based scenario. The agents are context-aware as agent actions are based upon sensor information, social information, and the behavior of co-agents.

Cardinal and Rank Ordering of Criteria — Addressing Prescription within Weight Elicitation

Weight elicitation methods in multi-criteria decision analysis (MCDA) are often cognitively demanding, require too much precision, time and effort. Some of the issues may be remedied by connecting elicitation methods to an inference engine facilitating a quick and easy method for decision-makers to use weaker input statements, yet being able to utilize these statements in a method for decision evaluation. In this paper, we propose a fast and practically useful weight elicitation method, answering to many of the requirements. The method builds on the ideas of rank-order methods, but can also take imprecise cardinal information into account. The method is subsequently employed in two real-life case studies and compared to a case where a simple ratio weight procedure using exact input statements was employed.

Speech emotion recognition in emotional feedbackfor Human-Robot Interaction

For robots to plan their actions autonomously and interact with people, recognizing human emotions is crucial. For most humans nonverbal cues such as pitch, loudness, spectrum, speech rate are effi ...

Reputation, inequality and meeting techniques: visualising user hierarchy to support collaboration

Equality within groups is ordinarily taken for granted when technology for e-democracy is conceived and developed. However, inequality in online communication is just as common as in other social contexts. Therefore, we have developed a groupware with the express purpose of illuminating imbalance of power. Inequalities are measured and made visible to users of the system, and they change dynamically as actions are taken by users. The system is based on democratic meeting techniques and is reminiscent of a strategy game based on social media. Each participant’s score within the game is dynamically calculated and reflects that user’s activity, others’ reactions to that activity and reactions to others’ activities. The calculations and weighing mechanisms are open to inspection and change by the users, and hierarchical roles reflecting game levels may be attached to system rights belonging to individual users and user groups. The prototype we present stems from the question of how to conceive of groupware based on diversity and is the result of combining social theory with algorithms for modelling and visualising user hierarchy and status. Empirical user tests suggest improvements to the prototype’s interface, which will be implemented and further evaluated by embedding the algorithms in a system for e-participation.