Logo des Repositoriums

Künstliche Intelligenz 31(2) - Mai 2017

Autor*innen mit den meisten Dokumenten  

Auflistung nach:

Neueste Veröffentlichungen

1 - 10 von 17
  • Zeitschriftenartikel
    Indoor Wayfinding: Interview with Christoph Hölscher and Ruth Conroy Dalton
    (KI - Künstliche Intelligenz: Vol. 31, No. 2, 2017) Krukar, Jakub; Hölscher, Christoph; Conroy Dalton, Ruth
  • Zeitschriftenartikel
    The path of least resistance
    (KI - Künstliche Intelligenz: Vol. 31, No. 2, 2017) Müller, Manuel; Ohm, Christina; Schwappach, Florin; Ludwig, Bernd
    Several studies show that pedestrians do not simply take the shortest route to their destination. In this paper, we address the question, which other factors influence their route decisions. We present a landmark based pedestrian navigation system that guides users through indoor and outdoor areas and forms the foundation of our research regarding route preferences. To investigate the issue which basic preferences exist, a qualitative pre-study was conducted $$(\mathrm{N} = 21).$$(N=21). In the main study we examined if routes chosen by participants in realistic scenarios deviate from the shortest possible route. With this, we explored if preferences, found in the pre-study, really influence the selection of routes $$(\mathrm{N} = 121).$$(N=121). In the pre-study most participants stated that they want to take the shortest route. In addition to that, it was frequently mentioned that entities along the route, like escalators, elevators, crowded areas, or unsafe areas should be preferred or avoided. The main study revealed that the participants, which were all familiar with the surroundings, take detours in indoor parts compared to the shortest route. Moreover, the participants avoided elevators and routes that lead through cafeterias and lecture halls.
  • Zeitschriftenartikel
    Space for Opinions
    (KI - Künstliche Intelligenz: Vol. 31, No. 2, 2017) Wrede, Britta
  • Zeitschriftenartikel
    (KI - Künstliche Intelligenz: Vol. 31, No. 2, 2017)
  • Zeitschriftenartikel
    Landmark Extraction from Web-Harvested Place Descriptions
    (KI - Künstliche Intelligenz: Vol. 31, No. 2, 2017) Kim, Junchul; Vasardani, Maria; Winter, Stephan
    Large corpora of place descriptions provide abundant human spatial knowledge, different from the geometry-based information stored in current GIS. These place descriptions, used in everyday communication, frequently refer to landmarks. This paper suggests a model for extracting landmarks from web-harvested place descriptions, considering the landmark’s cognitive significance. The model allows landmarks to be extracted according to different contexts via web harvesting and text classification methods. In this work, an implementation of our approach is used to extract context-based landmarks for a target area—Melbourne in Australia.
  • Zeitschriftenartikel
    Identifying Landmark Candidates Beyond Toy Examples
    (KI - Künstliche Intelligenz: Vol. 31, No. 2, 2017) Richter, Kai-Florian
    Incorporating references to landmarks in navigation systems requires having data on potential landmarks in the first place. While there have been many approaches in the scientific literature for identifying landmark candidates, these have hardly been picked up in actual, running systems. One major obstacle for this to happen may be that most—if not all—approaches presented so far are not scalable due to their underlying data requirements. In this paper, I will critically discuss existing approaches in light of their scalability. I will then suggest a way forward to more scalable solutions by combining in a smart way aspects of different approaches.
  • Zeitschriftenartikel
    Towards Efficiently Implementing Dodgson’s Formally Intractable Voting Rule
    (KI - Künstliche Intelligenz: Vol. 31, No. 2, 2017) Recknagel, Arne; Besold, Tarek R.
    Conflict of interest is the permanent companion of any population of agents (computational or biological). For that reason, the ability to compromise is of paramount importance, making voting a key element of societal mechanisms. A voting procedure often discussed in the literature and, due to its intuitiveness, also conceptually quite appealing is Charles Dodgson’s scoring rule, basically using the respective closeness to being a Condorcet winner for evaluating competing alternatives. In this paper, we offer insights into the practical limits of algorithms computing the exact Dodgson scores from a number of votes. While the problem itself is theoretically intractable, this work proposes and analyses five different solutions which try distinct approaches to practically solve the issue in an effective manner. Additionally, three of the discussed procedures can be run in parallel which has the potential of drastically improving computational performance on the problem.
  • Zeitschriftenartikel
    Off-Screen Landmarks on Mobile Devices: Levels of Measurement and the Perception of Distance on Resized Icons
    (KI - Künstliche Intelligenz: Vol. 31, No. 2, 2017) Li, Rui; Zhao, Jiayan
    While bringing portability and convenience to their users, the small screen size of mobile devices raises the concern that it might impact a user’s acquisition of spatial knowledge. Visualizing information of off-screen objects on mobile device has thus been introduced as a possible way to overcome this problem. Some approaches encode the distance to off-screen objects very well, but they have not considered the identities of objects, which could serve as easily-recognizable landmarks of recognition. Other approaches have addressed the visualization of distant objects’ identities as landmarks, but they have not considered the representation of distance to their actual locations. Following these approaches, this study introduces the use of visual variable size in the design of symbols for off-screen landmarks to translate both information about both direction and distance. To further investigate the efficiency of using these graduated size symbols, we apply ratio and ordinal levels of measurement to assign size to the symbols. Results show, size at the ordinal level leads to higher efficiency in understanding distance to off-screen locations. Both designs, however, yield challenges in participants’ understanding of distance based on the symbol’s size. As the initial step of investigating the use of visual variables in the design of symbols for off-screen landmarks, we suggest more visual variables be considered in follow-up designs to provide a more comprehensive understanding regarding the effectiveness of visualizing off-screen landmarks on mobile devices.
  • Zeitschriftenartikel
    A Cognitive Observer-Based Landmark-Preference Model
    (KI - Künstliche Intelligenz: Vol. 31, No. 2, 2017) Röser, Florian
  • Zeitschriftenartikel
    Dissertation Abstract: Empirically Measuring Salience of Objects for Use in Pedestrian Navigation
    (KI - Künstliche Intelligenz: Vol. 31, No. 2, 2017) Kattenbeck, Markus