http://nova.newcastle.edu.au/vital/access/services/Feed ${session.getAttribute("locale")} 5 http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:12629 Research Doctorate - Computer Science 2013-03-12T06:00:01.666Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:11631 Research Doctorate - Doctor of Philosophy (PhD) 2012-10-15T02:50:07.845Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:10064 This study describes a new approach for pedestrian behaviour analysis in simulated urban environments. A software system was developed to analyse the dynamics of pedestrians with a focus on their movement trajectories and the angle between the pedestrian's movement vector and their gaze vector. One-class support vector machines and dynamic time warping were applied for outlier detection in order to indentify noticeable changes in the visual behavior of individuals in a group of simulated pedestrians who walk past visually attractive objects. Results of simulation experiments demonstrate how different pedestrian behaviour characteristics can be detected and distinguished in a variety of abstract urban design scenarios. The described software system was designed to allow future real world applications by feeding video recordings of real pedestrians into the model. 2012-02-16T01:40:05.926Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:1378 This paper investigates the requirements of product modelling in the construction industry. Product models incorporate multifaceted aspects of design information (required at each stage of the lifecycle of buildings) by integrating additional information (such as time, costs, etc.) into a three-dimensional (3D) computer model, thereby adding intelligence to it. The project also investigates methodologies for automatically linking construction processes with 3D CAD models to allow users to visualize and simulate construction methodologies. Our study adopts a qualitative approach where semi-structured interviews were conducted with 11 key design and construction professionals from two major Australian companies. Data were coded in relation to six main clusters – themes and summaries of results are presented as repertory grids. The paper identifies some of the risks and opportunities of implementing nD modelling in the construction industry. Analysis of data indicates a shift to 3D CAD, with a strong interest being identified for integration of CAD and construction programming. Although the use of product models is not presently seen as feasible for this purpose, the increasing use of 3D CAD is seen as positive. Results indicate a need for alignment models and user-friendly technologies if product models are to assist communication between clients, consultants and construction companies. 2012-01-25T02:10:02.109Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:6077 Research Doctorate - Doctor of Philosophy (PhD) 2011-12-13T01:00:02.057Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:9283 This work presents a study on the genetic profile of the left and right hemispheres of the brain of a mouse model of Parkinson's disease (PO). The goal is to characterize, in a genetic basis, PO as a disease that affects these two brain regions in different ways. Using the same whole-genome microarray expression data introduced by Brown et al., we could find significant differences in the expression of some key genes, well-known to be involved in the mechanisms of dopamine production control and PD. The problem of selecting such genes was modeled as the MIN (a,/3)-FEATURE SET problem; a similar approach to that employed previously to find biomarkers for different types of cancer using gene expression microarray data. The Feature Selection method produced a series of genetic signatures for PO, with distinct expression profiles in the Parkinson's model and control mice experiments. In addition, a close examination of the genes composing those signatures shows that many of them belong to genetic pathways or have ontology annotations considered to be involved in the onset and development of PD. Such elements could provide new clues on which mechanisms are implicated in hemisphere differentiation in PD. 2011-11-08T23:00:04.328Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:8599 Evolving factor analysis (EFA), as the name implies, is a particular variant of factor analysis (FA). There are different versions of FA and they serve several different purposes; some of them are presented in detail in this monograph. One immediate and important information revealed by FA is the rank of the matrix of data that are analyzed. This rank is an indication of the complexity of the process represented by the data. Often it can be understood as the number of components in the system, but as we will see later, this is not a rule; often it is better to see the rank as related to the number of processes followed by the measurements. In mathematical terms, the definition of rank is straightforward – it is the number of linearly independent rows/columns. In chemistry, there is a considerable variety of situations and no clear and general statements can be made; we then use the expression ‘chemical rank’. 2011-08-10T06:10:03.999Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:8586 Model-based nonlinear data fitting is an invaluable instrument in the hand of any scientist, who is interested in the quantitative analysis of measured data. The range of analyses is wide and includes simple straight line fits to the global analysis of series of measurements of complex chemical processes. Data fitting comprises three main aspects. The first and central component of any data analysis is the collection of measured data that wait to be analysed. The second component of data fitting is a model that is used to quantitatively describe the data. The third aspect is the actual fitting routine, which determines the most likely values for the parameters of the model. 2011-08-10T06:00:07.395Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:6841 This paper investigates two different intelligent techniques—the neural network (NN) method and the simulated annealing (SA) algorithm for solving the inverse problem of Rutherford backscattering (RBS) with noisy data. The RBS inverse problem is to determine the sample structure information from measured spectra, which can be defined as either a function approximation or a non-linear optimization problem. Early studies emphasized on numerical methods and empirical fitting. In this work, we have applied intelligent techniques and compared their performance and effectiveness for spectral data analysis by solving the inverse problem. Since each RBS spectrum may contain up to 512 data points, principal component analysis is used to make the feature extraction so as to ease the complexity of constructing the network. The innovative aspects of our work include introducing dimensionality reduction and noise modeling. Experiments on RBS spectra from SiGe thin films on a silicon substrate show that the SA is more accurate but the NN is faster, though both methods produce satisfactory results. Both methods are resilient to 10% Poisson noise in the input. These new findings indicate that in RBS data analysis the NN approach should be preferred when fast processing is required; whereas the SA method becomes the first choice should the analysis accuracy be targeted. 2010-12-03T01:10:03.302Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:6631 This chapter illustrates the use of the combinatorial optimization models presented in Chapter 19 for the Feature Set selection and Gene Ordering problems to find genetic signatures for diseases using microarray data. We demonstrate the quality of this approach by using a microarray dataset from a mouse model of Parkinson's disease. The results are accompanied by a description of the currently known molecular functions and biological processes of the genes in the signatures. 2010-09-10T01:30:09.044Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:6628 The aim of this chapter is to present combinatorial optimization models and techniques for the analysis of microarray datasets. The chapter illustrates the application of a novel objective function that guides the search for high-quality solutions for sequential ordering of expression profiles. The approach is unsupervised and a metaheuristic method (a memetic algorithm) is used to provide high-quality solutions. For the problem of selecting discriminative groups of ienes, we used a supervised method that has provided good results in a variety of datasets. This chapter illustrates the application of these models in an Alzheimer's disease microarray dataset. 2010-09-10T01:30:06.283Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:6615 We present in this chapter the combined use of several recently introduced methodologies for the analysis of microarray datasets. These computational techniques are varied in type and very powerful when combined. We have selected a prostate cancer dataset which is available in the public domain to allow for further comparisons with existing methods. The task is to identify biomarkers that correlate with the clinical phenotype of interest, i.e., Gleason patterns 3,4, and 5. A supervised method, based on the mathematical formalism of (αβ)-k-feature sets, is used to select differentially expressed genes. After these "molecular signatures" are identified, we applied an unsupervised method (a memetic algorithm) to order the samples. The objective is to maximize a global measure of correlation in the two-dimensional display of gene expression profiles. With the resulting ordering and taxonomy we are able to identify samples that have been assigned a certain Gleason pattern, and have gene expression patterns different from most of the other samples in the group. We reiterate the approach to obtain molecular signatures that produce coherent patterns of gene expression in each of the three Gleason pattern groups, and we analyze the statistically significant patterns of gene expression that seem to be implicated in these different stages of disease. 2010-09-10T01:00:13.671Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:2378 2010-04-27T06:32:07.399Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:1657 Understanding your data through exploratory data analysis is a necessary first stage of data analysis particularly for observational data. The checking of data integrity and understanding the distributions, correlations and relationships between potentially important variables is a fundamental part of the analysis process prior to model development and hypothesis testing. In this paper, exploratory data analysis is used to assess the potential of laser return type and return intensity as variables for classification of individual trees or forest stands according to species. For narrow footprint lidar instruments that record up to two return amplitudes for each output pulse, the usual pre-classification of return data into first and last intensity returns camouflages the fact that a number of the return signals have only “single amplitude” (singular) returns. The importance of singular returns for species discrimination has received little discussion in the remote sensing literature. A map view of the different types of returns overlaid on field species data indicated that it is possible to visually distinguish between vegetation types that produce a high proportion of singular returns, compared to vegetation types that produce a lower proportion of singular returns, at least when using a specific laser footprint size. Using lidar data and the corresponding field data derived from a subtropical woodland area of South East Queensland, Australia, map scatterplots of return types combined with field data enabled, in some cases, visual discrimination at the individual tree level between White Cypress Pine (Callitris glaucophylla) and Poplar Box (Eucalyptus populnea). While a clear distinction between these two species was not always visually obvious at the individual tree level, due to other extraneous sources of variation in the dataset, the observation was supported in general at the site level. Sites dominated by Poplar Box generally exhibited a lower proportion of singular returns compared to sites dominated by Cypress Pine. While return intensity statistics for this particular dataset were not found to be as useful for classification as the proportions of laser return types, an examination of the return intensity data leads to an explanation of how return intensity statistics are affected by forest structure. Exploratory data analysis indicated that a large component of variation in the intensity of the return signals from a forest canopy is associated with reflections of only part of the laser footprint. Consequently, intensity return statistics for the forest canopy, such as average and standard deviation, are related not only to the reflective properties of the vegetation, but also to the larger scale properties of the forest such as canopy openness and the spacing and type of foliage components within individual tree crowns. 2010-04-27T06:28:12.977Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:482 Understanding your data through exploratory data analysis is a necessary first stage of data analysis particularly for observational data. The checking of data integrity and understanding the distributions, correlations and relationships between potentially important variables is a fundamental part of the analysis process prior to model development and hypothesis testing. In this paper, exploratory data analysis is used to assess the potential of laser return type and return intensity as variables for classification of individual trees or forest stands according to species. For narrow footprint lidar instruments that record up to two return amplitudes for each output pulse, the usual preclassification of return data into first and last intensity returns camouflages the fact that a number of the return signals have only "single amplitude' (singular) returns. The importance of singular returns for species discrimination has received little discussion in the remote sensing literature. A map view of the different types of returns overlaid on field species data indicated that it is possible to visually distinguish between vegetation types that produce a high proportion of singular returns, compared to vegetation types that produce a lower proportion of singular returns, at least when using a specific laser footprint size. Using lidar data and the corresponding field data derived from a subtropical woodland area of South East Queensland, Australia, map scatterplots of return types combined with field data enabled, in some cases, visual discrimination at the individual tree level between White Cypress Pine (Callitris glaucophylla) and Poplar Box (Eucalyptus populnea). While a clear distinction between these two species was not always visually obvious at the individual tree level, due to other extraneous sources of variation in the dataset, the observation was supported in general at the site level. Sites dominated by Poplar Box generally exhibited a lower proportion of singular returns compared to sites dominated by Cypress Pine. While return intensity statistics for this particular dataset were not found to be as useful for classification as the proportions of laser return types, an examination of the return intensity data leads to an explanation of how return intensity statistics are affected by forest structure. Exploratory data analysis indicated that a large component of variation in the intensity of the return signals from a forest canopy is associated with reflections of only part of the laser footprint. Consequently, intensity return statistics for the forest canopy, such as average and standard deviation, are related not only to the reflective properties of the vegetation, but also to the larger scale properties of the forest such as canopy openness and the spacing and type of foliage components within individual tree crowns. 2010-04-27T05:48:02.249Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:4079 This paper investigates the needs and requirements for product modelling in the construction industry. A product (or nD) model refers to an extension of the traditional "building information model". This model incorporates multi-faceted aspects of design information (required at each stage of the lifecycle of a building) by integrating additional information (such as time, costs and safety codes) into a three dimensional (3D) computer model, thereby adding intelligence to it The study adopts a qualitative approach where semi-structured interviews were conducted with 11 key design and construction professionals from two major Australian companies. Data were coded in relation to six main clusters themes and summaries of results are presented as repertory grids. The paper identifies risks v opportunities of implementing nD modelling in the construction industry. Results show the industry's shift to 3D computer aided design (CAD), with a strong interest being identified for CAD, programming and scheduling integration. Although the use of product models is not seen as feasible at this point in time, the increasing use of 3D CAD is seen as a positive step. Furthermore results indicate a need for alignment models and user-friendly technologies if product models are to assist communication between clients, consultants and construction companies. 2010-04-27T04:56:06.037Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:3222 Merging spectroscopic imaging and chemometrics enhances the outcomes of instrumental technology and data analysis. Multivariate exploratory and resolution methods can be adapted to image analysis and provide global and local information about pure compounds in an imaged sample. Knowing in detail how the chemical compounds are distributed over the scanned surface gives valuable information about essential issues in the manufacture and the characterization of products, such as evenness of composition and, therefore, homogeneity of the sample. The power to detect and to locate impurities is also greatly enhanced because these unwanted compounds could show locally large concentrations (and signals), even though their abundance on the surface is very low. The capabilities of this combination are shown in an example of pharmaceutical product control, where analysis of the end product requires chemical characterization and quantitative information at global and local levels. The approach used and the kind of information obtained is general and can be applied to the analysis of images in other fields. 2010-04-27T04:53:21.920Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:4461 The study by Drs Psaty and Kronmal and the accompanying Editorial by Drs DeAngelis and Fontanarosa regarding rofecoxib illustrate the recurring problem of discretionary or data-driven analysis. Given the possibility of selective analysis based on observed data and the risk of positive results due to chance alone, it is critical to know in detail which analyses were prespecified and when the prespecification occurred. This can only be done via a full protocol repository. With no protocol or a timeline documentation of its provenance, these details are never made public so that it is not possible to see how published analyses compare with protocol-specified analyses. This transparency is needed to be confident in the validity of the results. 2010-04-27T04:52:05.892Z ]]> http://nova.newcastle.edu.au/vital/access/manager/Repository/uon:5330 Purpose: The purpose of this paper is to compare data collected by automated form processing with manual data collection for clinical indicators (CIs) in paediatric emergency medicine. Design/methodology/approach: Paediatric patients presenting with croup, asthma, bronchiolitis, head injury and gastroenteritis in August 2006 were identified by ICD 9 coding and a traditional manual audit was performed by two data collectors. Data were collected on a total of 16 CIs for these five illnesses. Manual audit data were then compared to information collected for this same patient population using TELEform, an automated forms processing (AFP) system that had been employed for over two years. Findings: Teleform data were only available for 24 patients compared to information for 127 patients identified by ICD 9 coding and manual audit. Teleform data overestimated compliance with clinical guidelines by 17 percent giving an overall departmental agreement with CIs of 90.6 percent compared to 73.5 percent in the manual audit. Additionally, manual audit demonstrated that when the clinical guideline was incorporated into the clinical record, compliance was 92.5 percent compared to 51.3 percent when it was not. Originality/value: This single center study demonstrates that data collected by AFP such as TELEform, overestimate emergency department performance regarding CIs compliance. Departments that use automated data collection tools need to establish relationships between such data and data collected via more traditional auditing methods. 2010-04-27T04:30:43.903Z ]]>