Human vs. machine : identification of bat species from their echolocation calls by humans and by artificial neural networks

Automated remote ultrasound detectors allow large amounts of data on bat presence and activity to be collected. Processing of such data involves identifying bat species from their echolocation calls. Automated species identification has the potential to provide more consistent, predictable, and potentially higher levels of accuracy than identification by humans. In contrast, identification by humans permits flexibility and intelligence in identification, as well as the incorporation of features and patterns that may be difficult to quantify. We compared humans with artificial neural networks (ANNs) in their ability to classify short recordings of bat echolocation calls of variable signal to noise ratios; these sequences are typical of those obtained from remote automated recording systems that are often used in large-scale ecological studies. We presented 45 recordings (1–4 calls) produced by known species of bats to ANNs and to 26 human participants with 1 month to 23 years of experience in acoustic identification of bats. Humans correctly classified 86% of recordings to genus and 56% to species; ANNs correctly identified 92% and 62%, respectively. There was no significant difference between the performance of ANNs and that of humans, but ANNs performed better than about 75% of humans. There was little relationship between the experience of the human participants and their classification rate. However, humans with <1 year of experience performed worse than others. Currently, identification of bat echolocation calls by humans is suitable for ecological research, after careful consideration of biases. However, improvements to ANNs and the data that they are trained on may in future increase their performance to beyond those demonstrated by humans.

Further development of discrete computational techniques for calculation of restricted diffusion propagators in porous media

Magnetic resonance is a well-established tool for structural characterisation of porous media. Features of pore-space morphology can be inferred from NMR diffusion-diffraction plots or the time-dependence of the apparent diffusion coefficient. Diffusion NMR signal attenuation can be computed from the restricted diffusion propagator, which describes the distribution of diffusing particles for a given starting position and diffusion time. We present two techniques for efficient evaluation of restricted diffusion propagators for use in NMR porous-media characterisation. The first is the Lattice Path Count (LPC). Its physical essence is that the restricted diffusion propagator connecting points A and B in time t is proportional to the number of distinct length-t paths from A to B. By using a discrete lattice, the number of such paths can be counted exactly. The second technique is the Markov transition matrix (MTM). The matrix represents the probabilities of jumps between every pair of lattice nodes within a single timestep. The propagator for an arbitrary diffusion time can be calculated as the appropriate matrix power. For periodic geometries, the transition matrix needs to be defined only for a single unit cell. This makes MTM ideally suited for periodic systems. Both LPC and MTM are closely related to existing computational techniques: LPC, to combinatorial techniques; and MTM, to the Fokker-Planck master equation. The relationship between LPC, MTM and other computational techniques is briefly discussed in the paper. Both LPC and MTM perform favourably compared to Monte Carlo sampling, yielding highly accurate and almost noiseless restricted diffusion propagators. Initial tests indicate that their computational performance is comparable to that of finite element methods. Both LPC and MTM can be applied to complicated pore-space geometries with no analytic solution. We discuss the new methods in the context of diffusion propagator calculation in porous materials and model biological tissues.

Automatic UAV forced landing site detection using machine learning

The commercialization of aerial image processing is highly dependent on the platforms such as UAVs (Unmanned Aerial Vehicles). However, the lack of an automated UAV forced landing site detection system has been identified as one of the main impediments to allow UAV flight over populated areas in civilian airspace. This article proposes a UAV forced landing site detection system that is based on machine learning approaches including the Gaussian Mixture Model and the Support Vector Machine. A range of learning parameters are analysed including the number of Guassian mixtures, support vector kernels including linear, radial basis function Kernel (RBF) and polynormial kernel (poly), and the order of RBF kernel and polynormial kernel. Moreover, a modified footprint operator is employed during feature extraction to better describe the geometric characteristics of the local area surrounding a pixel. The performance of the presented system is compared to a baseline UAV forced landing site detection system which uses edge features and an Artificial Neural Network (ANN) region type classifier. Experiments conducted on aerial image datasets captured over typical urban environments reveal improved landing site detection can be achieved with an SVM classifier with an RBF kernel using a combination of colour and texture features. Compared to the baseline system, the proposed system provides significant improvement in term of the chance to detect a safe landing area, and the performance is more stable than the baseline in the presence of changes to the UAV altitude.

Making robust decisions for conservation with restricted money and knowledge

1. In conservation decision-making, we operate within the confines of limited funding. Furthermore, we often assume particular relationships between management impact and our investment in management. The structure of these relationships, however, is rarely known with certainty - there is model uncertainty. We investigate how these two fundamentally limiting factors in conservation management, money and knowledge, impact optimal decision-making. 2. We use information-gap decision theory to find strategies for maximizing the number of extant subpopulations of a threatened species that are most immune to failure due to model uncertainty. We thus find a robust framework for exploring optimal decision-making. 3. The performance of every strategy decreases as model uncertainty increases. 4. The strategy most robust to model uncertainty depends not only on what performance is perceived to be acceptable but also on available funding and the time horizon over which extinction is considered. 5. Synthesis and applications. We investigate the impact of model uncertainty on robust decision-making in conservation and how this is affected by available conservation funding. We show that subpopulation triage can be a natural consequence of robust decision-making. We highlight the need for managers to consider triage not as merely giving up, but as a tool for ensuring species persistence in light of the urgency of most conservation requirements, uncertainty and the poor state of conservation funding. We illustrate this theory by a specific application to allocation of funding to reduce poaching impact on the Sumatran tiger Panthera tigris sumatrae in Kerinci Seblat National Park. © 2008 The Authors.

Machine learning approaches to analysing textual injury surveillance data: A systematic review

Objective To synthesise recent research on the use of machine learning approaches to mining textual injury surveillance data. Design Systematic review. Data sources The electronic databases which were searched included PubMed, Cinahl, Medline, Google Scholar, and Proquest. The bibliography of all relevant articles was examined and associated articles were identified using a snowballing technique. Selection criteria For inclusion, articles were required to meet the following criteria: (a) used a health-related database, (b) focused on injury-related cases, AND used machine learning approaches to analyse textual data. Methods The papers identified through the search were screened resulting in 16 papers selected for review. Articles were reviewed to describe the databases and methodology used, the strength and limitations of different techniques, and quality assurance approaches used. Due to heterogeneity between studies meta-analysis was not performed. Results Occupational injuries were the focus of half of the machine learning studies and the most common methods described were Bayesian probability or Bayesian network based methods to either predict injury categories or extract common injury scenarios. Models were evaluated through either comparison with gold standard data or content expert evaluation or statistical measures of quality. Machine learning was found to provide high precision and accuracy when predicting a small number of categories, was valuable for visualisation of injury patterns and prediction of future outcomes. However, difficulties related to generalizability, source data quality, complexity of models and integration of content and technical knowledge were discussed. Conclusions The use of narrative text for injury surveillance has grown in popularity, complexity and quality over recent years. With advances in data mining techniques, increased capacity for analysis of large databases, and involvement of computer scientists in the injury prevention field, along with more comprehensive use and description of quality assurance methods in text mining approaches, it is likely that we will see a continued growth and advancement in knowledge of text mining in the injury field.

Detection of rain in acoustic recordings of the environment using machine learning techniques

This thesis is concerned with the detection and prediction of rain in environmental recordings using different machine learning algorithms. The results obtained in this research will help ecologists to efficiently analyse environmental data and monitor biodiversity.

Robot control of a 3,500 tonne mining machine

The mining industry is highly suitable for the application of robotics and automation technology since the work is arduous, dangerous and often repetitive. This paper describes the development of an automation system for a physically large and complex field robotic system - a 3,500 tonne mining machine (a dragline). The major components of the system are discussed with a particular emphasis on the machine/operator interface. A very important aspect of this system is that it must work cooperatively with a human operator, seamlessly passing the control back and forth in order to achieve the main aim - increased productivity.

Attentive visual tracking

The research reported here addresses the problem of detecting and tracking independently moving objects from a moving observer in real time, using corners as object tokens. Local image-plane constraints are employed to solve the correspondence problem removing the need for a 3D motion model. The approach relaxes the restrictive static-world assumption conventionally made, and is therefore capable of tracking independently moving and deformable objects. The technique is novel in that feature detection and tracking is restricted to areas likely to contain meaningful image structure. Feature instantiation regions are defined from a combination of odometry informatin and a limited knowledge of the operating scenario. The algorithms developed have been tested on real image sequences taken from typical driving scenarios. Preliminary experiments on a parallel (transputer) architecture indication that real-time operation is achievable.

The ghost in the machine: Legal challenges of neural interface devices

Neural interface devices and the melding of mind and machine, challenge the law in determining where civil liability for injury, damage or loss should lie. The ability of the human mind to instruct and control these devices means that in a negligence action against a person with a neural interface device, determining the standard of care owed by him or her will be of paramount importance. This article considers some of the factors that may influence the court’s determination of the appropriate standard of care to be applied in this situation, leading to the conclusion that a new standard of care might evolve.

Applications of real-time machine vision to the control of mining equipment

The mining industry presents us with a number of ideal applications for sensor based machine control because of the unstructured environment that exists within each mine. The aim of the research presented here is to increase the productivity of existing large compliant mining machines by retrofitting with enhanced sensing and control technology. The current research focusses on the automatic control of the swing motion cycle of a dragline and an automated roof bolting system. We have achieved: * closed-loop swing control of an one-tenth scale model dragline; * single degree of freedom closed-loop visual control of an electro-hydraulic manipulator in the lab developed from standard components.

Producing the NAPLAN Machine: A Schizoanalytic Cartography

Everything revolves around desiring-machines and the production of desire… Schizoanalysis merely asks what are the machinic, social and technical indices on a socius that open to desiring-machines (Deleuze & Guattari, 1983, pp. 380-381). Achievement tests like NAPLAN are fairly recent, yet common, education policy initiatives in much of the Western world. They intersect with, use and change pre-existing logics of education, teaching and learning. There has been much written about the form and function of these tests, the ‘stakes’ involved and the effects of their practice. This paper adopts a different “angle of vision” to ask what ‘opens’ education to these regimes of testing(Roy, 2008)? This paper builds on previous analyses of NAPLAN as a modulating machine, or a machine characterised by the increased intensity of connections and couplings. One affect can be “an existential disquiet” as “disciplinary subjects attempt to force coherence onto a disintegrating narrative of self”(Thompson & Cook, 2012, p. 576). Desire operates at all levels of the education assemblage, however our argument is that achievement testing manifests desire as ‘lack’; seen in the desire for improved results, the desire for increased control, the desire for freedom, the desire for acceptance to name a few. For Deleuze and Guattari desire is irreducible to lack, instead desire is productive. As a productive assemblage, education machines operationalise and produce through desire; “Desire is a machine, and the object of the desire is another machine connected to it”(Deleuze & Guattari, 1983, p. 26). This intersection is complexified by the strata at which they occur, the molar and molecular connections and flows they make possible. Our argument is that when attention is paid to the macro and micro connections, the machines built and disassembled as a result of high-stakes testing, a map is constructed that outlines possibilities, desires and blockages within the education assemblage. This schizoanalytic cartography suggests a new analysis of these ‘axioms’ of testing and accountability. It follows the flows and disruptions made possible as different or altered connections are made and as new machines are brought online. Thinking of education machinically requires recognising that “every machine functions as a break in the flow in relation to the machine to which it is connected, but at the same time is also a flow itself, or the production of flow, in relation to the machine connected to it”(Deleuze & Guattari, 1983, p. 37). Through its potential to map desire, desire-production and the production of desire within those assemblages that have come to dominate our understanding of what is possible, Deleuze and Guattari’s method of schizoanalysis provides a provocative lens for grappling with the question of what one can do, and what lines of flight are possible.

Lateralization of temporal lobe epilepsy based on resting-state functional magnetic resonance imaging and machine learning

Lateralization of temporal lobe epilepsy (TLE) is critical for successful outcome of surgery to relieve seizures. TLE affects brain regions beyond the temporal lobes and has been associated with aberrant brain networks, based on evidence from functional magnetic resonance imaging. We present here a machine learning-based method for determining the laterality of TLE, using features extracted from resting-state functional connectivity of the brain. A comprehensive feature space was constructed to include network properties within local brain regions, between brain regions, and across the whole network. Feature selection was performed based on random forest and a support vector machine was employed to train a linear model to predict the laterality of TLE on unseen patients. A leave-one-patient-out cross validation was carried out on 12 patients and a prediction accuracy of 83% was achieved. The importance of selected features was analyzed to demonstrate the contribution of resting-state connectivity attributes at voxel, region, and network levels to TLE lateralization.

A penalty-based genetic algorithm for the migration cost-aware virtual machine placement problem in cloud data centers

In the past few years, the virtual machine (VM) placement problem has been studied intensively and many algorithms for the VM placement problem have been proposed. However, those proposed VM placement algorithms have not been widely used in today's cloud data centers as they do not consider the migration cost from current VM placement to the new optimal VM placement. As a result, the gain from optimizing VM placement may be less than the loss of the migration cost from current VM placement to the new VM placement. To address this issue, this paper presents a penalty-based genetic algorithm (GA) for the VM placement problem that considers the migration cost in addition to the energy-consumption of the new VM placement and the total inter-VM traffic flow in the new VM placement. The GA has been implemented and evaluated by experiments, and the experimental results show that the GA outperforms two well known algorithms for the VM placement problem.

Harnessing information from injury narrative in the 'big data' era: Understanding and applying machine learning for injury surveillance

Objective Vast amounts of injury narratives are collected daily and are available electronically in real time and have great potential for use in injury surveillance and evaluation. Machine learning algorithms have been developed to assist in identifying cases and classifying mechanisms leading to injury in a much timelier manner than is possible when relying on manual coding of narratives. The aim of this paper is to describe the background, growth, value, challenges and future directions of machine learning as applied to injury surveillance. Methods This paper reviews key aspects of machine learning using injury narratives, providing a case study to demonstrate an application to an established human-machine learning approach. Results The range of applications and utility of narrative text has increased greatly with advancements in computing techniques over time. Practical and feasible methods exist for semi-automatic classification of injury narratives which are accurate, efficient and meaningful. The human-machine learning approach described in the case study achieved high sensitivity and positive predictive value and reduced the need for human coding to less than one-third of cases in one large occupational injury database. Conclusion The last 20 years have seen a dramatic change in the potential for technological advancements in injury surveillance. Machine learning of ‘big injury narrative data’ opens up many possibilities for expanded sources of data which can provide more comprehensive, ongoing and timely surveillance to inform future injury prevention policy and practice.

Design and Development of an In-Vehicle Human Machine Interface to Improve Fuel Efficiency & Safety

Road transport plays a significant role in various industries and mobility services around the globe and has a vital impact on our daily lives. However it also has serious impacts on both public health and the environment. In-vehicle feedback systems are a relatively new approach to encouraging driver behaviour change for improving fuel efficiency and safety in automotive environments. While many studies claim that the adoption of eco-driving practices, such as eco-driving training programs and in-vehicle feedback to drivers, has the potential to improve fuel efficiency, limited research has integrated safety and eco-driving. Therefore, this research seeks to use human factors related theories and practices to inform the design and evaluation of an in-vehicle Human Machine Interface (HMI) providing real-time driver feedback with the aim of improving both fuel efficiency and safety.