» DK » R » RODIL.DK

Total: 25

Choose link from "Titles, links and description words view":

Or switch to "Titles and links view".
  • Virtual Savannah
    target groups of the zoo â the school children We describe the first tentative steps to substitute the facilitating teachersâ â pushâ learning approach with an interactive virtual world for children to explore hence using game design and game dynamics as a bridging tool between subject and children To gain insights into how the children use and navigate the Virtual Savannah we have concurrently prototyped a framework for logging user information as quantifiable data In comparison with traditional observation techniques the logging system can show a hidden corpus of data which is inspected from techniques visualizing it as information visualization Subsequently we provide design approaches for a further use of the logging tool and share insights into the data handling pipeline We provide details behind the analysis design implementation and testing of the system in situ on 4 th grade children The results detail findings of the systemâ s potential as a functional prototype for conveying information and the results show that children with low prior knowledge on the subject of African ecology have improved knowledge one week after the use of Virtual Savannah if compared with their prior knowledge  The articles below detail the corpus of results from the data logging component and videos pictures from the system Publications will be once updated located in their own section FUNCTIONALITY AND DATA PROCESSING Report excerption The Virtual Savannah VS is developed in Unity3D and runs as a built executable To quantifiably inspect users of our system we have considered the use of screen recording software through i e Camtasia but previous experiences with this software showed a heavy drop in system performance which creates a paradox in the development where we are forced to reduce graphical fidelity to be able to run a measuring tool to investigate use of the system hence also implicitly measuring on visual quality Another argument for investing time and effort in developing a tool inside our running code is that screen recordings only deliver data in sense of videos which also require qualitative evaluation techniques thus not producing quantifiable data We approached the problem by writing already available camera coordinates in the system into xml files from inside our program which removed human error factors such as manually writing concurrently as the system runs and it also opened possibilities to log large corpora of data with very little stress on the hardware software I e we logged camera coordinates from 23 users at our public presentation described in the section about public pilot evaluation and in total our system logged 11 016 coordinate pairs which when stored in a file only consumed 152kb of storage space This makes it easy to handle while running our system but due to its file size it also opens possibilities to be executed on smaller systems such as phones or to be run online After ensuring robustness we decided to include other parameters to log which include all our animals the temporal length of interaction the spatial position

    Original URL path: (2015-09-25)
    Open archived version from archive