http://dl.gi.de/handle/20.500.12116/5017 2026-07-22T20:33:32Z 2026-07-22T20:33:32Z Neumayr, Thomas Augstein, Mirjam http://dl.gi.de/handle/20.500.12116/5114 2017-11-15T15:01:28Z 2012-01-01T00:00:00Z

Recommender Systems: Between Acceptance and Refusal Neumayr, Thomas; Augstein, Mirjam Augstein, Mirjam; Heckmann, Dominik; Herder, Eelco Recommender Systems (RSs) are a prominent solution to the problem of information overload on the web. It is impossible for users to process or even understand all information presented to them. Also, it becomes more and more difficult for an individual to identify appropriate concrete pieces of information or information sources. RSs aim at adapting the presented content and the order in which it is presented to users’ individual needs, based on their preferences and past behavior. Yet, a system can only provide accurate recommendations if it has been authentically used before, i.e., been able to collect information about a user. As authentic usage depends on a user’s acceptance, the success of RSs in general is strongly dependent on acceptance also. If recommendations seem inappropriate, the trust in the system will fade. This paper presents a study analyzing how and to what extent different factors like transparency or controllability influence acceptance in the context of web-based recommendation.

2012-01-01T00:00:00Z Ilina, Elena Abel, Fabian Houben, Geert-Jan http://dl.gi.de/handle/20.500.12116/5116 2017-11-15T15:01:28Z 2012-01-01T00:00:00Z

Mining Twitter for Cultural Patterns Ilina, Elena; Abel, Fabian; Houben, Geert-Jan Augstein, Mirjam; Heckmann, Dominik; Herder, Eelco Adaptive applications rely on the knowledge of their users, their needs and differences. For instance, in the scope of the ImReal 1 project, a training process is adapted to users’ origins using information on user cultural backgrounds. For inferring culture-specific information from available microblogging content, we monitor the usage of Twitter elements such as hashtags, web links and user mentions. We analyze how users from different cultural groups employ these elements when they tweet. This allows us to get insights on microblogging patterns for different cultural groups of Twitter users and an outlook into user preferences and traits towards sharing content with others, time preferences, and social networking attitudes. Potentially, such information can be used for adapting software applications in accord with user culture-specific behavioral traits.

2012-01-01T00:00:00Z Kurschl, Werner Augstein, Mirjam Stitz, Holger http://dl.gi.de/handle/20.500.12116/5115 2017-11-15T15:01:28Z 2012-01-01T00:00:00Z

Adaptive User Interfaces on Tablets to Support People With Disabilities Kurschl, Werner; Augstein, Mirjam; Stitz, Holger Augstein, Mirjam; Heckmann, Dominik; Herder, Eelco With the advent of tablet computers, touch screens, gesture-based interaction and speech recognition, sophisticated applications with Natural User Interfaces (NUIs) become state of the art. NUIs have the potential to support people with disabilities, e.g., in their daily activities or in acquiring specific skills. Yet, one main challenge is that this user group has diverse abilities and handicaps so that an interaction design must be highly configurable to make NUIs beneficial. The introduction of adaptivity might be promising in order to overcome configuration complexity and effort. This paper presents an approach to adaptive user interfaces on tablets to support people with disabilities.

2012-01-01T00:00:00Z Opasjumruskit, Kobkaew Expósito, Jesús König-Ries, Birgitta Nauerz, Andreas Welsch, Martin http://dl.gi.de/handle/20.500.12116/5113 2017-11-15T15:01:28Z 2012-01-01T00:00:00Z

MERCURY: User Centric Device and Service Processing – Demo paper Opasjumruskit, Kobkaew; Expósito, Jesús; König-Ries, Birgitta; Nauerz, Andreas; Welsch, Martin Augstein, Mirjam; Heckmann, Dominik; Herder, Eelco In this paper, we present MERCURY, a platform for simple, user-centric integration and management of heterogeneous devices and services via a web-based interface. In contrast to existing approaches, MERCURY is geared towards non-IT-savvy end users. It enables these end users to easily interconnect devices, which can act as sensors or actuators, to model rules that trigger actions. Sets of rules allow users to model entire, often reoccurring, scenarios. Also, these must be user-centric and context adaptive. It shall thus enable users to take full advantage of the potential for support in everyday life such integration offers. Technically, our solution is based on Portal technology. We describe a tangible scenario to portray the steps a user will need to take to achieve the desired functionality.

2012-01-01T00:00:00Z