Refine
Year of publication
- 2010 (42) (remove)
Document Type
- Conference Proceeding (21)
- Contribution to a Periodical (6)
- Article (reviewed) (5)
- Article (unreviewed) (4)
- Part of a Book (2)
- Patent (2)
- Book (1)
- Master's Thesis (1)
Conference Type
- Konferenzartikel (16)
- Konferenz-Abstract (3)
- Konferenz-Poster (1)
- Sonstiges (1)
Language
- English (42) (remove)
Keywords
- E-Learning (4)
- Brennstoffzelle (2)
- Elektrode (2)
- Glucose (2)
- mobile learning (2)
- 3D virtual reality (1)
- Additive Manufacture (1)
- Aflatoxin (1)
- Agent <Künstliche Intelligenz> (1)
- Agent Fitness (1)
Institute
Open Access
- Open Access (17)
- Closed Access (11)
- Closed (7)
- Bronze (5)
Principles of Animation Film
(2010)
In their famous work on prospect theory Kahneman and Tversky have presented a couple of examples where human decision making deviates from rational decision making as defined by decision theory. This paper describes the use of extended behavior networks to model human decision making in the sense of prospect theory. We show that the experimental findings of non-rational decision making described by Kahneman and Tversky can be reproduced using a slight variation of extended behavior networks.
This paper gives an overview of the implementation of an Active Noise Control system on the TMS320C6713 Digital Signal Processor from Texas Instruments in the Digital Signal Processing Lab at Hochschule Offenburg, Germany. This system is implemented considering some non-ideal environmental conditions on a real system instead of being limited to computer simulations. Changes over time on the physical acoustical path as well as reverberation and variation on the power of the reference signal can strongly degrade the performance of the system or even lead to instability. In order to try to minimize these effects, the Active Noise Control system was designed to support a fast and easy implementation and evaluation of different algorithms on the DSP in real-time. In Section 1 a brief introduction about active noise control system is given and in section 2 the basic algorithm is described. In section 3 the implementation of the system is described and in section 4 some final considerations are given.
This Master's Thesis discusses intelligent sensor networks considering autonomous sensor placement strategies and system health management. Sensor networks for an intelligent system design process have been researched recently. These networks consist of a distributed collective of sensing units, each with the abilities of individual sensing and computation. Such systems can be capable of self-deployment and must be scalable, long-lived and robust. With distributed sensor networks, intelligent sensor placement for system design and online system health management are attractive areas of research. Distributed sensor networks also cause optimization problems, such as decentralized control, system robustness and maximization of coverage in a distributed system. This also includes the discovery and analysis of points of interest within an environment. The purpose of this study was to investigate a method to control sensor placement in a world with several sources and multiple types of information autonomously. This includes both controlling the movement of sensor units and filtering of the gathered information depending on individual properties to increase system performance, defined as a good coverage. Additionally, online system health management was examined in this study regarding the case of agent failures and autonomous policy reconfiguration if sensors are added to or removed from the system. Two different solution strategies were devised, one where the environment was fully observable, and one with only partial observability. Both strategies use evolutionary algorithms based on artificial neural networks for developing control policies. For performance measurement and policy evaluation, different multiagent objective functions were investigated. The results of the study show that in the case of a world with multiple types of information, individual control strategies performed best because of their abilities to control the movement of a sensor entity and to filter the sensed information. This also includes system robustness in case of sensor failures where other sensing units must recover system performance. Additionally, autonomous policy reconfiguration after adding or removing of sensor agents was successful. This highlights that intelligent sensor agents are able to adapt their individual control policies considering new circumstances.
Prof. Gitte Lindgaard, from the University of Carleton, Canada, says that viewing only some milliseconds of the first page of a website defines our general opinion about it [1]. For an online-shop, it would therefore be essential to have a first page that is not only pleasing to the eye, but also understandable enough to not loose the attention of the user. More and more companies are nowadays using the Internet not only as a showcase anymore, but as a full-strength selling tool, needing thus to convince their users and clients at first glance. This paper shows the analysis of two online-shops in the magazines’ field thanks to eye-tracking. With the analysis of the testers’ glances and their comments during and after the test, the usability of these two websites has been evaluated.
After approximately 200 years, a comprehensive access to the texts of Humboldt’s extraordinary exploration of the Americans is within sight. To open the legacy to the public for free access the Humboldt Digital Library (HDL) project has been developing a dynamic amount of data related to studies of Alexander von Humboldt. The library includes a range of texts, tables and images, as well as many tools that assist mining the data and navigating the system.
The idea of this game is to use a flashcard system to create a short story in a foreign language. The story is developed by a group of participants by exchanging sentences via a flashcard system. This way the participants can learn from each other by knowledge sharing without fear of making mistakes because the group members are anonymous. Moreover they do not need a constant support from a teacher.
The developed solution enables the presentation of animations and 3D virtual reality (VR) on mobile devices and is well suited for mobile learning, thus creating new possibilities in the area of e-learning worldwide. Difficult relations in physics as well as intricate experiments in optics can be visualised on mobile devices without need for a personal computer.