Database and narratological representation of Australian Aboriginal knowledge as information visualisation using a game engine

Current database technologies do not support contextualised representations of multi-dimensional narratives. This paper outlines a new approach to this problem using a multi-dimensional database served in a 3D game environment. Preliminary results indicate it is a particularly efficient method for the types of contextualised narratives used by Australian Aboriginal peoples to tell their stories about their traditional landscapes and knowledge practices. We discuss the development of a tool that complements rather than supplants direct experience of these traditional knowledge practices.

Representation and learning of visual information for pose recognition

Recovering position from sensor information is an important problem in mobile robotics, known as localisation. Localisation requires a map or some other description of the environment to provide the robot with a context to interpret sensor data. The mobile robot system under discussion is using an artificial neural representation of position. Building a geometrical map of the environment with a single camera and artificial neural networks is difficult. Instead it would be simpler to learn position as a function of the visual input. Usually when learning images, an intermediate representation is employed. An appropriate starting point for biologically plausible image representation is the complex cells of the visual cortex, which have invariance properties that appear useful for localisation. The effectiveness for localisation of two different complex cell models are evaluated. Finally the ability of a simple neural network with single shot learning to recognise these representations and localise a robot is examined.

Intertidal habitat conservation: identifying conservation targets in the absence of detailed biological information

1. Growing concern associated with threats to the marine environment has resulted in an increased demand for marine reserves that conserve representative and adequate examples of biodiversity. Often, the decisions about where to locate reserves must be made in the absence of detailed information on the patterns of distribution of the biota. Alternative approaches are required that include defining habitats using surrogates for biodiversity. Surrogate measures of biodiversity enable decisions about where to locate marine reserves to be made more reliably in the absence of detailed data on the distribution of species. 2. Intertidal habitat types derived using physical properties of the shoreline were used as a surrogate for intertidal biodiversity to assist with the identification of sites for inclusion in a candidate system of intertidal marine reserves for 17 463 km of the mainland coast of Queensland, Australia. This represents the first systematic approach, on essentially one-dimensional data, using fine-scale (tens to hundreds of metres) intertidal habitats to identify a system of marine reserves for such a large length of coast. A range of solutions would provide for the protection of a representative example of intertidal habitats in Queensland. 3. The design and planning of marine and terrestrial protected areas systems should not be undertaken independently of each other because it is likely to lead to inadequate representation of intertidal habitats in either system. The development of reserve systems specially designed to protect intertidal habitats should be integrated into the design of terrestrial and marine protected area systems. Copyright (c) 2005 John Wiley & Sons, Ltd.

Effectiveness of document representation for classification

Conventionally, document classification researches focus on improving the learning capabilities of classifiers. Nevertheless, according to our observation, the effectiveness of classification is limited by the suitability of document representation. Intuitively, the more features that are used in representation, the more comprehensive that documents are represented. However, if a representation contains too many irrelevant features, the classifier would suffer from not only the curse of high dimensionality, but also overfitting. To address this problem of suitableness of document representations, we present a classifier-independent approach to measure the effectiveness of document representations. Our approach utilises a labelled document corpus to estimate the distribution of documents in the feature space. By looking through documents in this way, we can clearly identify the contributions made by different features toward the document classification. Some experiments have been performed to show how the effectiveness is evaluated. Our approach can be used as a tool to assist feature selection, dimensionality reduction and document classification.