Growing self-organizing map for online continuous clustering

The internet age has fuelled an enormous explosion in the amount of information generated by humanity. Much of this information is transient in nature, created to be immediately consumed and built upon (or discarded). The field of data mining is surprisingly scant with algorithms that are geared towards the unsupervised knowledge extraction of such dynamic data streams. This chapter describes a new neural network algorithm inspired by self-organising maps. The new algorithm is a hybrid algorithm from the growing self-organising map (GSOM) and the cellular probabilistic self-organising map (CPSOM). The result is an algorithm which generates a dynamically growing feature map for the purpose of clustering dynamic data streams and tracking clusters as they evolve in the data stream.

Introducing personal liability under the continuous disclosure regime : The essentials and non-essentials

Under the Federal Government's CLERP 9 legislation, expected at the time of writing to come into force in July 2004, personal liability will be introduced for the first time under the continuous disclosure regime. Individuals who are 'involved' in a failure to immediately disclose materially price sensitive information to the market will be subject to a civil penalty, in addition to the company being liable. According to the author, the introduction of personal liability per se is not contentious and indeed is a favourable change; what is questionable, however, is whether 'involvement' in a contravention is the appropriate test for imposing personal liability in relation to breaches of the continuous disclosure provisions. Based on the case law to date on the meaning of 'involved', there is particular uncertainty as to whether an individual would need to have actual knowledge that non-disclosed information is 'materially price sensitive' in order to satisfy the test of 'involved' in the context of continuous disclosure, or whether mere knowledge that the information has not been disclosed would be sufficient. This uncertainty arises due to the vague concept of 'essential matters' which the courts have developed as a test for what degree of knowledge a person needs to have in order to be 'involved'. The author argues that all the confusion as to what 'involved' means could be addressed by removing the word 'essential' from the dialogue, so that the test of 'involvement' would simply be based on whether the particular person had actual knowledge of each of the factual elements constituting the offence.

Western maps/Yanyuwa meaning: an interview with John Bradley

In July 2003 an important one-volume text, Forget about Flinders: A Yanyuwa atlas of the south west gulf of Carpentaria produced in a limited edition of 14 copies, returned to Yanyuwa country and to the families who collaborated with John Bradley and artist Nona Cameron on the project. Subsequently, a second edition of 20 copies has been released, mainly to institutions. It is the most comprehensive attempt yet to restore Yanyuwa names to country and to produce a multilayered, dynamic, history-rich, and bilingual representation of how country is known in this community, and how the central song cycle texts intersect with Yanyuwa tradition. What follows is a condensed and edited interview with Frances Devlin-Glass, in which John Bradley discusses the motivations, the hybridised methodologies employed, the innovations of this new genre, and the pedagogical ends served by this latest iteration of Yanyuwa song cycles.

A multi-wavelet based technique for calculating dense 2D disparity maps from stereo

A vision based approach for calculating accurate 3D models of the objects is presented. Generally industrial visual inspection systems capable of accurate 3D depth estimation rely on extra hardware tools like laser scanners or light pattern projectors. These tools improve the accuracy of depth estimation but also make the vision system costly and cumbersome. In the proposed algorithm, depth and dimensional accuracy of the produced 3D depth model depends on the existing reference model instead of the information from extra hardware tools. The proposed algorithm is a simple and cost effective software based approach to achieve accurate 3D depth estimation with minimal hardware involvement. The matching process uses the well-known coarse to fine strategy, involving the calculation of matching points at the coarsest level with consequent refinement up to the finest level. Vector coefficients of the wavelet transform-modulus are used as matching features, where wavelet transform-modulus maxima defines the shift invariant high-level features with phase pointing to the normal of the feature surface. The technique addresses the estimation of optimal corresponding points and the corresponding 2D disparity maps leading to the creation of accurate depth perception model.

Developing and presenting conceptual maps of course content

Teaching online involves providing an environment that is interactive and engaging. A large part of this is providing suitable learning resources. In this talk we will demonstrate an efficient method for producing conceptual maps of the actual course content, showing the structure of the subject for students in a visual way. The structures that result allows for learning resources to be linked in as required. The maps are developed using PowerPoint but they can be deployed in a web-friendly format or on CD-ROM.

Conductive wool yarns by continuous vapour phase polymerization of pyrrole

Conducting polypyrrole (PPy) coated wool yarns were prepared by a continuous vapour polymerization technique, using a speed of 1 m/min with different iron(III) chloride (FeCl₃) as the oxidant at different concentrations. The resistivities, tensile properties, longitudinal and cross-sectional views of PPy-coated wool yarns were investigated. Optimum specific electrical resistances of 2.96 Ω g/cm² at 80 g/L FeCl₃ and 1.69 Ω g/cm² at 70 g/L FeCl₃ were obtained for 500 and 400 twist per meter (TPM) yarns, respectively. PPy-coated wool yarns exhibited higher elongation than uncoated yarns. Longitudinal and cross-sectional views of the yarns indicate that PPy coating penetrated deep into the yarn cross-section and a uniform coating was obtained on the surface of the yarn surface.