Stephan Hammer

Trust-based decision-making for smart and adaptive environments

Smart environments are able to support users during their daily life. For example, smart energy systems can be used to support energy saving by controlling devices, such as lights or displays, depending on context information, such as the brightness in a room or the presence of users. However, proactive decisions should also match the users’ preferences to maintain the users’ trust in the system. Wrong decisions could negatively influence the users’ acceptance of a system and at worst could make them abandon the system. In this paper, a trust-based model, called User Trust Model (UTM), for automatic decision-making is proposed, which is based on Bayesian networks. The UTM’s construction, the initialization with empirical data gathered in an online survey, and its integration in an office setting are described. Furthermore, the results of a live study and a live survey analyzing the users’ experience and acceptance are presented.

The Neighbor-Trust Metric to Measure Reputation in Organic Computing Systems

Reputation is an important aspect of trust. If no direct trust experiences are available, one needs to rely on reputation data from other sources. In this paper we present the Neighbor-Trust metric that exploits these communication capabilities of a network by directly asking all neighbors of a target communication partner for reputation trust data. This results in a reputation path of length one, but also in a vulnerability to attacks by unknown, lying entities that try to promote not trustworthy entities. However, by adding weights for reputation data given by entities and a learning mechanism the Neighbor-Trust metric is able to identify and adapt to lying participants in the network by reducing the weight their reputation data has in future reputation calculations. We present an evaluation for the metric and show how to exclude lying participants from the network.

Exploring Digital Image Frames for Lifestyle Intervention to Improve Well-being of Older Adults

This contribution addresses the development of technology for senior users with the aim to improve their general wellbeing. We present a prototype system named CARE that is used for in-situ testing in a seniors home and combines functionality of a digital image frame with an active recommender mode. The purpose of the recommender is to provide the user with context-specific recommendations. Recommendations are chosen on the basis of sensor data and a well-being model to carefully decide on at which point in time what kind of activity will be most suitable to suggest.

Personalization of Content on Public Displays Driven by the Recognition of Group Context

Personalization of content on public displays relies on the knowledge of spectator interests and real-time recognition of social context. In busy public places, with numerous individuals circulating daily, the knowledge of individual interests becomes unrealistic. This paper presents an approach for automatic personalization which, instead of individual profiles, relies on group context. The system recognizes the constellation of spectators in front of a public display, based on their disposition and gender. Thus, the approach provides an important prerequisite for a completely automated personalization, requiring no input from the spectator side, neither for training, nor for real-time content adaptation. The experiment conducted in a public area showed that the presented approach can successfully identify the differences in the content observation of various groups. Moreover, the approach provides an insight into the diversity of circulating groups, and gives a hint about spectators’ emotional and conversational response to the content.

Trust-based Decision-making for the Adaptation of Public Displays in Changing Social Contexts

Public displays may adapt intelligently to the social context, tailoring information on the screen, for example, to the profiles of spectators, their gender or based on their mutual proximity. However, such adaptation decisions should on the one hand match user preferences and on the other maintain the user’s trust in the system. A wrong decision can negatively influence the user’s acceptance of a system, cause frustration and, as a result, make users abandon the system. In this paper, we propose a trust-based mechanism for automatic decision-making, which is based on Bayesian Networks. We present the process of network construction, initialization with empirical data, and validation. The validation demonstrates that the mechanism generates accurate decisions on adaptation which match user preferences and support user trust.

The Automatic Trust Management of Self-Adaptive Multi-Display Environments

This paper presents an approach to automatically manage user trust in self-adaptive ubiquitous computing systems which grounds on a context interpreter and a Bayesian Network as well as a feedback control loop that also provides solutions for a system adaptation. Providing knowledge to the automatic trust management of self-adaptive ubiquitous systems is of special interest due to the fact that during the use of these systems situations appear which can impair user trust and thus user acceptance. Some of these situations and adaptation approaches are presented in this paper.

Location-Based interaction with children for edutainment

Our mixed-reality installation features two cooperating characters that integrate multiple users by location-based tracking into the interaction, allowing for dynamic storylines.

Investigating Politeness Strategies and Their Persuasiveness for a Robotic Elderly Assistant

This work is targeted towards the development of a Robotic Elderly Assistant REA system that provides assistance in the form of recommendations to support single-living elderly people in their domestic environment. To avoid potential face threats the REA should be as polite as possible whilst keeping a certain persuasiveness to promote its recommendations. This paper investigates different verbalizations of the REAs recommendations regarding their perceived politeness as well as their persuasiveness. We present the results of a laboratory study with younger adults and a user study with the inhabitants of a retirement home. Results suggest that the different politeness strategies reflected different levels of politeness in both studies, while their perceived persuasiveness needs further investigation in the domain of elderly care.

A User Trust Model for Automatic Decision-Making in Ubiquitous and Self-Adaptive Environments

Ubiquitous Environments are able to support users during their daily life by intelligently self-adapting to changed contexts. Examples include home automation systems which can support energy saving by switching off unused devices or public displays which enable users to present and interact with data, but maintain the users’ privacy by hiding sensible data if others pass by. However, such proactive adaptations could also cause frustration and thus harm the users’ acceptance and trust towards a system if they do not match the users’ preferences or are not self-explanatory. In the worst case, wrong or incomprehensible decisions by the system even could make the users abandon the system. To address this concern, we propose a generic trust-based model, called User Trust Model (UTM), which facilitates automatic decision-making in ubiquitous and self-adaptive environments. It is supposed to monitor users’ trust in the system and to select context-aware system actions that maintain, restore, or even foster user trust. In this chapter, the construction of the generic model as well as its integration into two case studies will be presented. We will provide a detailed description of how to customise the UTM for the respective scenarios and share results and experiences from various studies conducted with the developed systems.

Trust-Based Decision-Making for Energy-Aware Device Management

Smart energy systems are able to support users in saving energy by controlling devices, such as lights or displays, depending on context information, such as the brightness in a room or the presence of users. However, proactive decisions should also match the users’ preferences to maintain users’ trust in the system. Wrong decisions could negatively influence users’ acceptance of a system and at worst could make them abandon the system. In this paper, a trust-based model, called User Trust Model (UTM), for automatic decision-making is proposed, which is based on Bayesian Networks. The UTM’s construction, the initialization with empirical data gathered in an online survey, and its integration in an office setting are described. Furthermore, the results of a user study investigating users’ experience and acceptance are presented.