A Novel Nonparametric Density Estimator

We present a novel nonparametric density estimator and a new data-driven bandwidth selection method with excellent properties. The approach is in- spired by the principles of the generalized cross entropy method. The pro- posed density estimation procedure has numerous advantages over the tra- ditional kernel density estimator methods. Firstly, for the first time in the nonparametric literature, the proposed estimator allows for a genuine incor- poration of prior information in the density estimation procedure. Secondly, the approach provides the first data-driven bandwidth selection method that is guaranteed to provide a unique bandwidth for any data. Lastly, simulation examples suggest the proposed approach outperforms the current state of the art in nonparametric density estimation in terms of accuracy and reliability.

A two-phase approach for detecting recombination in nucleotide sequences

Genetic recombination can produce heterogeneous phylogenetic histories within a set of homologous genes. Delineating recombination events is important in the study of molecular evolution, as inference of such events provides a clearer picture of the phylogenetic relationships among different gene sequences or genomes. Nevertheless, detecting recombination events can be a daunting task, as the performance of different recombination-detecting approaches can vary, depending on evolutionary events that take place after recombination. We recently evaluated the effects of post-recombination events on the prediction accuracy of recombination-detecting approaches using simulated nucleotide sequence data. The main conclusion, supported by other studies, is that one should not depend on a single method when searching for recombination events. In this paper, we introduce a two-phase strategy, applying three statistical measures to detect the occurrence of recombination events, and a Bayesian phylogenetic approach in delineating breakpoints of such events in nucleotide sequences. We evaluate the performance of these approaches using simulated data, and demonstrate the applicability of this strategy to empirical data. The two-phase strategy proves to be time-efficient when applied to large datasets, and yields high-confidence results.

Migrating eprints.org data to a Fez repository

This document records the process of migrating eprints.org data to a Fez repository. Fez is a Web-based digital repository and workflow management system based on Fedora (http://www.fedora.info/). At the time of migration, the University of Queensland Library was using EPrints 2.2.1 [pepper] for its ePrintsUQ repository. Once we began to develop Fez, we did not upgrade to later versions of eprints.org software since we knew we would be migrating data from ePrintsUQ to the Fez-based UQ eSpace. Since this document records our experiences of migration from an earlier version of eprints.org, anyone seeking to migrate eprints.org data into a Fez repository might encounter some small differences. Moving UQ publication data from an eprints.org repository into a Fez repository (hereafter called UQ eSpace (http://espace.uq.edu.au/) was part of a plan to integrate metadata (and, in some cases, full texts) about all UQ research outputs, including theses, images, multimedia and datasets, in a single repository. This tied in with the plan to identify and capture the research output of a single institution, the main task of the eScholarshipUQ testbed for the Australian Partnership for Sustainable Repositories project (http://www.apsr.edu.au/). The migration could not occur at UQ until the functionality in Fez was at least equal to that of the existing ePrintsUQ repository. Accordingly, as Fez development occurred throughout 2006, a list of eprints.org functionality not currently supported in Fez was created so that programming of such development could be planned for and implemented.

A simple data logger for student-designed rocket experiments.

The final-year project for Mechanical & Space Engineering students at UQ often involves the design and flight testing of an experiment. This report describes the design and use of a simple data logger that should be suitable for collecting data from the students' flight experiments. The exercise here was taken as far as the construction of a prototype device that is suitable for ground-based testing, say, the static firing of a hybrid rocket motor.

Bimanual aiming and overt attention: one law for two hands

Reaching to interact with an object requires a compromise between the speed of the limb movement and the required end-point accuracy. The time it takes one hand to move to a target in a simple aiming task can be predicted reliably from Fitts' law, which states that movement time is a function of a combined measure of amplitude and accuracy constraints (the index of difficulty, ID). It has been assumed previously that Fitts' law is violated in bimanual aiming movements to targets of unequal ID. We present data from two experiments to show that this assumption is incorrect: if the attention demands of a bimanual aiming task are constant then the movements are well described by a Fitts' law relationship. Movement time therefore depends not only on ID but on other task conditions, which is a basic feature of Fitts' law. In a third experiment we show that eye movements are an important determinant of the attention demands in a bimanual aiming task. The results from the third experiment extend the findings of the first two experiments and show that bimanual aiming often relies on the strategic co-ordination of separate actions into a seamless behaviour. A number of the task specific strategies employed by the adult human nervous system were elucidated in the third experiment. The general strategic pattern observed in the hand trajectories was reflected by the pattern of eye movements recorded during the experiment. The results from all three experiments demonstrate that eye movements must be considered as an important constraint in bimanual aiming tasks.

Constraints on the spatiotemporal accuracy of interceptive action: effects of target size on hitting a moving object

Results of two experiments are reported that examined how people respond to rectangular targets of different sizes in simple hitting tasks. If a target moves in a straight line and a person is constrained to move along a linear track oriented perpendicular to the targetrsquos motion, then the length of the target along its direction of motion constrains the temporal accuracy and precision required to make the interception. The dimensions of the target perpendicular to its direction of motion place no constraints on performance in such a task. In contrast, if the person is not constrained to move along a straight track, the targetrsquos dimensions may constrain the spatial as well as the temporal accuracy and precision. The experiments reported here examined how people responded to targets of different vertical extent (height): the task was to strike targets that moved along a straight, horizontal path. In experiment 1 participants were constrained to move along a horizontal linear track to strike targets and so target height did not constrain performance. Target height, length and speed were co-varied. Movement time (MT) was unaffected by target height but was systematically affected by length (briefer movements to smaller targets) and speed (briefer movements to faster targets). Peak movement speed (Vmax) was influenced by all three independent variables: participants struck shorter, narrower and faster targets harder. In experiment 2, participants were constrained to move in a vertical plane normal to the targetrsquos direction of motion. In this task target height constrains the spatial accuracy required to contact the target. Three groups of eight participants struck targets of different height but of constant length and speed, hence constant temporal accuracy demand (different for each group, one group struck stationary targets = no temporal accuracy demand). On average, participants showed little or no systematic response to changes in spatial accuracy demand on any dependent measure (MT, Vmax, spatial variable error). The results are interpreted in relation to previous results on movements aimed at stationary targets in the absence of visual feedback.

Hitting a moving target: Perception and action in the timing of rapid interceptions

Different interceptive tasks and modes of interception (hitting or capturing) do not necessarily involve similar control processes. Control based on preprogramming of movement parameters is possible for actions with brief movement times but is now widely rejected; continuous perceptuomotor control models are preferred for all types of interception. The rejection of preprogrammed control and acceptance of continuous control is evaluated for the timing of rapidly executed, manual hitting actions. It is shown that a preprogrammed control model is capable of providing a convincing account of observed behavior patterns that avoids many of the arguments that have been raised against it. Prominent continuous perceptual control models are analyzed within a common framework and are shown to be interpretable as feedback control strategies. Although these models can explain observations of on-line adjustments to movement, they offer only post hoc explanations for observed behavior patterns in hitting tasks and are not directly supported by data. It is proposed that rapid manual hitting tasks make up a class of interceptions for which a preprogrammed strategy is adopted-a strategy that minimizes the role of visual feedback. Such a strategy is effective when the task demands a high degree of temporal accuracy.

Equational Reasoning as a Tool for Data Analysis

A combination of deductive reasoning, clustering, and inductive learning is given as an example of a hybrid system for exploratory data analysis. Visualization is replaced by a dialogue with the data.

Leadership Attributes and Cultural Values in Australia and New Zealand Compared: An Initial Report Based on GLOBE Data.

This paper reports a comparative study of Australian and New Zealand leadership attributes, based on the GLOBE (Global Leadership and Organizational Behavior Effectiveness) program. Responses from 344 Australian managers and 184 New Zealand managers in three industries were analyzed using exploratory and confirmatory factor analysis. Results supported some of the etic leadership dimensions identified in the GLOBE study, but also found some emic dimensions of leadership for each country. An interesting finding of the study was that the New Zealand data fitted the Australian model, but not vice versa, suggesting asymmetric perceptions of leadership in the two countries.

Selection bias in gene extraction on the basis of microarray gene-expression data

In the context of cancer diagnosis and treatment, we consider the problem of constructing an accurate prediction rule on the basis of a relatively small number of tumor tissue samples of known type containing the expression data on very many (possibly thousands) genes. Recently, results have been presented in the literature suggesting that it is possible to construct a prediction rule from only a few genes such that it has a negligible prediction error rate. However, in these results the test error or the leave-one-out cross-validated error is calculated without allowance for the selection bias. There is no allowance because the rule is either tested on tissue samples that were used in the first instance to select the genes being used in the rule or because the cross-validation of the rule is not external to the selection process; that is, gene selection is not performed in training the rule at each stage of the cross-validation process. We describe how in practice the selection bias can be assessed and corrected for by either performing a cross-validation or applying the bootstrap external to the selection process. We recommend using 10-fold rather than leave-one-out cross-validation, and concerning the bootstrap, we suggest using the so-called. 632+ bootstrap error estimate designed to handle overfitted prediction rules. Using two published data sets, we demonstrate that when correction is made for the selection bias, the cross-validated error is no longer zero for a subset of only a few genes.

Sampling phylogenetic tree space with the generalized Gibbs sampler

The generalized Gibbs sampler (GGS) is a recently developed Markov chain Monte Carlo (MCMC) technique that enables Gibbs-like sampling of state spaces that lack a convenient representation in terms of a fixed coordinate system. This paper describes a new sampler, called the tree sampler, which uses the GGS to sample from a state space consisting of phylogenetic trees. The tree sampler is useful for a wide range of phylogenetic applications, including Bayesian, maximum likelihood, and maximum parsimony methods. A fast new algorithm to search for a maximum parsimony phylogeny is presented, using the tree sampler in the context of simulated annealing. The mathematics underlying the algorithm is explained and its time complexity is analyzed. The method is tested on two large data sets consisting of 123 sequences and 500 sequences, respectively. The new algorithm is shown to compare very favorably in terms of speed and accuracy to the program DNAPARS from the PHYLIP package.

A two-phase strategy for detecting recombination in nucleotide sequences

Genetic recombination can produce heterogeneous phylogenetic histories within a set of homologous genes. Delineating recombination events is important in the study of molecular evolution, as inference of such events provides a clearer picture of the phylogenetic relationships among different gene sequences or genomes. Nevertheless, detecting recombination events can be a daunting task, as the performance of different recombination-detecting approaches can vary, depending on evolutionary events that take place after recombination. We previously evaluated the effects of post-recombination events on the prediction accuracy of recombination-detecting approaches using simulated nucleotide sequence data. The main conclusion, supported by other studies, is that one should not depend on a single method when searching for recombination events. In this paper, we introduce a two-phase strategy, applying three statistical measures to detect the occurrence of recombination events, and a Bayesian phylogenetic approach to delineate breakpoints of such events in nucleotide sequences. We evaluate the performance of these approaches using simulated data, and demonstrate the applicability of this strategy to empirical data. The two-phase strategy proves to be time-efficient when applied to large datasets, and yields high-confidence results.

Special issue on advances in data mining and its applications

Data mining is the process to identify valid, implicit, previously unknown, potentially useful and understandable information from large databases. It is an important step in the process of knowledge discovery in databases, (Olaru & Wehenkel, 1999). In a data mining process, input data can be structured, seme-structured, or unstructured. Data can be in text, categorical or numerical values. One of the important characteristics of data mining is its ability to deal data with large volume, distributed, time variant, noisy, and high dimensionality. A large number of data mining algorithms have been developed for different applications. For example, association rules mining can be useful for market basket problems, clustering algorithms can be used to discover trends in unsupervised learning problems, classification algorithms can be applied in decision-making problems, and sequential and time series mining algorithms can be used in predicting events, fault detection, and other supervised learning problems (Vapnik, 1999). Classification is among the most important tasks in the data mining, particularly for data mining applications into engineering fields. Together with regression, classification is mainly for predictive modelling. So far, there have been a number of classification algorithms in practice. According to (Sebastiani, 2002), the main classification algorithms can be categorized as: decision tree and rule based approach such as C4.5 (Quinlan, 1996); probability methods such as Bayesian classifier (Lewis, 1998); on-line methods such as Winnow (Littlestone, 1988) and CVFDT (Hulten 2001), neural networks methods (Rumelhart, Hinton & Wiliams, 1986); example-based methods such as k-nearest neighbors (Duda & Hart, 1973), and SVM (Cortes & Vapnik, 1995). Other important techniques for classification tasks include Associative Classification (Liu et al, 1998) and Ensemble Classification (Tumer, 1996).

A framework for electricity price spike analysis with advanced data mining methods

There are many techniques for electricity market price forecasting. However, most of them are designed for expected price analysis rather than price spike forecasting. An effective method of predicting the occurrence of spikes has not yet been observed in the literature so far. In this paper, a data mining based approach is presented to give a reliable forecast of the occurrence of price spikes. Combined with the spike value prediction techniques developed by the same authors, the proposed approach aims at providing a comprehensive tool for price spike forecasting. In this paper, feature selection techniques are firstly described to identify the attributes relevant to the occurrence of spikes. A simple introduction to the classification techniques is given for completeness. Two algorithms: support vector machine and probability classifier are chosen to be the spike occurrence predictors and are discussed in details. Realistic market data are used to test the proposed model with promising results.

Appropriate surveying methods in the Philippines: Can modern GPS units help?

An investigation was undertaken to test the effectiveness of two procedures for recording boundaries and plot positions for scientific studies on farms on Leyte Island, the Philippines. The accuracy of a Garmin 76 Global Positioning System (GPS) unit and a compass and chain was checked under the same conditions. Tree canopies interfered with the ability of the satellite signal to reach the GPS and therefore the GPS survey was less accurate than the compass and chain survey. Where a high degree of accuracy is required, a compass and chain survey remains the most effective method of surveying land underneath tree canopies, providing operator error is minimised. For a large number of surveys and thus large amounts of data, a GPS is more appropriate than a compass and chain survey because data are easily up-loaded into a Geographic Information System (GIS). However, under dense canopies where satellite signals cannot reach the GPS, it may be necessary to revert to a compass survey or a combination of both methods.