A comparison of classification algorithms within the Classifynder pollen imaging system

We describe an investigation into how Massey University's Pollen Classifynder can accelerate the understanding of pollen and its role in nature. The Classifynder is an imaging microscopy system that can locate, image and classify slide based pollen samples. Given the laboriousness of purely manual image acquisition and identification it is vital to exploit assistive technologies like the Classifynder to enable acquisition and analysis of pollen samples. It is also vital that we understand the strengths and limitations of automated systems so that they can be used (and improved) to compliment the strengths and weaknesses of human analysts to the greatest extent possible. This article reviews some of our experiences with the Classifynder system and our exploration of alternative classifier models to enhance both accuracy and interpretability. Our experiments in the pollen analysis problem domain have been based on samples from the Australian National University's pollen reference collection (2890 grains, 15 species) and images bundled with the Classifynder system (400 grains, 4 species). These samples have been represented using the Classifynder image feature set. In addition to the Classifynder's native neural network classifier, we have evaluated linear discriminant, support vector machine, decision tree and random forest classifiers on these data with encouraging results. Our hope is that our findings will help enhance the performance of future releases of the Classifynder and other systems for accelerating the acquisition and analysis of pollen samples. © 2013 AIP Publishing LLC.

An evaluation of the neocognitron

We describe a sequence of experiments investigating the strengths and limitations of Fukushima's neocognitron as a handwritten digit classifier. Using the results of these experiments as a foundation, we propose and evaluate improvements to Fukushima's original network in an effort to obtain higher recognition performance. The neocognitron's performance is shown to be strongly dependent on the choice of selectivity parameters and we present two methods to adjust these variables. Performance of the network under the more effective of the two new selectivity adjustment techniques suggests that the network fails to exploit the features that distinguish different classes of input data. To avoid this shortcoming, the network's final layer cells were replaced by a nonlinear classifier (a multilayer perceptron) to create a hybrid architecture. Tests of Fukushima's original system and the novel systems proposed in this paper suggest that it may be difficult for the neocognitron to achieve the performance of existing digit classifiers due to its reliance upon the supervisor's choice of selectivity parameters and training data. These findings pertain to Fukushima's implementation of the system and should not be seen as diminishing the practical significance of the concept of hierarchical feature extraction embodied in the neocognitron. © 1997 IEEE.

A Boundary-Oriented Chinese Segmentation Method Using NGram Mutual Information

This paper describes our participation in the Chinese word segmentation task of CIPS-SIGHAN 2010. We implemented an n-gram mutual information (NGMI) based segmentation algorithm with the mixed-up features from unsupervised, supervised and dictionarybased segmentation methods. This algorithm is also combined with a simple strategy for out-of-vocabulary (OOV) word recognition. The evaluation for both open and closed training shows encouraging results of our system. The results for OOV word recognition in closed training evaluation were however found unsatisfactory.

Detecting rare events using Kullback-Leibler divergence

One main challenge in developing a system for visual surveillance event detection is the annotation of target events in the training data. By making use of the assumption that events with security interest are often rare compared to regular behaviours, this paper presents a novel approach by using Kullback-Leibler (KL) divergence for rare event detection in a weakly supervised learning setting, where only clip-level annotation is available. It will be shown that this approach outperforms state-of-the-art methods on a popular real-world dataset, while preserving real time performance.

A high performance liquid chromatography and electrospray ionization mass spectrometry method for the analysis of the natural medicine, Forsythia suspensa

Samples of Forsythia suspensa from raw (Laoqiao) and ripe (Qingqiao) fruit were analyzed with the use of HPLC-DAD and the EIS-MS techniques. Seventeen peaks were detected, and of these, twelve were identified. Most were related to the glucopyranoside molecular fragment. Samples collected from three geographical areas (Shanxi, Henan and Shandong Provinces), were discriminated with the use of hierarchical clustering analysis (HCA), discriminant analysis (DA), and principal component analysis (PCA) models, but only PCA was able to provide further information about the relationships between objects and loadings; eight peaks were related to the provinces of sample origin. The supervised classification models-K-nearest neighbor (KNN), least squares support vector machines (LS-SVM), and counter propagation artificial neural network (CP-ANN) methods, indicated successful classification but KNN produced 100% classification rate. Thus, the fruit were discriminated on the basis of their places of origin.

Analysing reflective text for learning analytics : an approach using anomaly recontextualisation

Reflective writing is an important learning task to help foster reflective practice, but even when assessed it is rarely analysed or critically reviewed due to its subjective and affective nature. We propose a process for capturing subjective and affective analytics based on the identification and recontextualisation of anomalous features within reflective text. We evaluate 2 human supervised trials of the process, and so demonstrate the potential for an automated Anomaly Recontextualisation process for Learning Analytics.

Herds of methane chambers grazing bubbles

Water to air methane emissions from freshwater reservoirs can be dominated by sediment bubbling (ebullitive) events. Previous work to quantify methane bubbling from a number of Australian sub-tropical reservoirs has shown that this can contribute as much as 95% of total emissions. These bubbling events are controlled by a variety of different factors including water depth, surface and internal waves, wind seiching, atmospheric pressure changes and water levels changes. Key to quantifying the magnitude of this emission pathway is estimating both the bubbling rate as well as the areal extent of bubbling. Both bubbling rate and areal extent are seldom constant and require persistent monitoring over extended time periods before true estimates can be generated. In this paper we present a novel system for persistent monitoring of both bubbling rate and areal extent using multiple robotic surface chambers and adaptive sampling (grazing) algorithms to automate the quantification process. Individual chambers are self-propelled and guided and communicate between each other without the need for supervised control. They can maintain station at a sampling site for a desired incubation period and continuously monitor, record and report fluxes during the incubation. To exploit the methane sensor detection capabilities, the chamber can be automatically lowered to decrease the head-space and increase concentration. The grazing algorithms assign a hierarchical order to chambers within a preselected zone. Chambers then converge on the individual recording the highest 15 minute bubbling rate. Individuals maintain a specified distance apart from each other during each sampling period before all individuals are then required to move to different locations based on a sampling algorithm (systematic or adaptive) exploiting prior measurements. This system has been field tested on a large-scale subtropical reservoir, Little Nerang Dam, and over monthly timescales. Using this technique, localised bubbling zones on the water storage were found to produce over 50,000 mg m-2 d-1 and the areal extent ranged from 1.8 to 7% of the total reservoir area. The drivers behind these changes as well as lessons learnt from the system implementation are presented. This system exploits relatively cheap materials, sensing and computing and can be applied to a wide variety of aquatic and terrestrial systems.

Autonomous hot metal carrier: Navigation and manipulation with a 20 tonne industrial vehicle

This paper reports work on the automation of a hot metal carrier, which is a 20 tonne forklift-type vehicle used to move molten metal in aluminium smelters. To achieve efficient vehicle operation, issues of autonomous navigation and materials handling must be addressed. We present our complete system and experiments demonstrating reliable operation. One of the most significant experiments was five-hours of continuous operation where the vehicle travelled over 8 km and conducted 60 load handling operations. Finally, an experiment where the vehicle and autonomous operation were supervised from the other side of the world via a satellite phone network are described.

Autonomous hot metal carrier

This paper reports work involved with the automation of a Hot Metal Carrier — a 20 tonne forklift-type vehicle used to move molten metal in aluminium smelters. To achieve efficient vehicle operation, issues of autonomous navigation and materials handling must be addressed. We present our complete system and experiments demontrating reliable operation. One of the most significant experiments was five-hours of continuous operation where the vehicle travelled over 8 km and conducted 60 load handling operations. We also describe an experiment where the vehicle and autonomous operation were supervised from the other side of the world via a satellite phone network.

An exercise intervention during chemotherapy for women with recurrent ovarian cancer: A feasibility study

Objective: The aim of this study was to determine the feasibility of a combined supervised and home-based exercise intervention during chemotherapy for women with recurrent ovarian cancer. Secondary aims were to determine the impact of physical activity on physical and psychological outcomes and on chemotherapy completion rates. Methods: Women with recurrent ovarian cancer were recruited from 3 oncology outpatient clinics in Sydney and Canberra, Australia. All participants received an individualized exercise program that consisted of 90 minutes or more of low to moderate aerobic, resistance, core stability, and balance exercise per week, for 12 weeks. Feasibility was determined by recruitment rate, retention rate, intervention adherence, and adverse events. Aerobic capacity, muscular strength, fatigue, sleep quality, quality of life, depression, and chemotherapy completion rates were assessed at weeks 0, 12, and 24. Results: Thirty participants were recruited (recruitment rate, 63%), with a retention rate of 70%. Participants averaged 196 ± 138 min · wk of low to moderate physical activity throughout the intervention, with adherence to the program at 81%. There were no adverse events resulting from the exercise intervention. Participants who completed the study displayed significant improvements in quality of life (P = 0.017), fatigue (P = 0.004), mental health (P = 0.007), muscular strength (P = 0.001), and balance (P = 0.003) after the intervention. Participants completing the intervention had a higher relative dose intensity than noncompleters (P = 0.03). Conclusions: A program consisting of low to moderate exercise of 90 min · wk was achieved by two-thirds of women with recurrent ovarian cancer in this study, with no adverse events reported. Randomized control studies are required to confirm the benefits of exercise reported in this study.

The psychosocial purpose of driving and its relationship with the risky driving behaviour of young novice drivers

Background The overrepresentation of young drivers in road crashes, injuries and fatalities around the world has resulted in a breadth of injury prevention efforts including education, enforcement, engineering, and exposure control. Despite multifaceted intervention, the young driver problem remains a challenge for injury prevention researchers, practitioners and policy-makers. The intractable nature of young driver crash risks suggests that a deeper understanding of their car use – that is, the purpose of their driving – is required to inform the design of more effective young driver countermeasures. Aims This research examined the driving purpose reported by young drivers, including the relationship with self-reported risky driving behaviours including offences. Methods Young drivers with a Learner or Provisional licence participated in three online surveys (N1 = 656, 17–20 years; N2 = 1051, 17–20 years; N3 = 351, 17–21 years) as part of a larger state-wide project in Queensland, Australia. Results A driving purpose scale was developed (the PsychoSocial Purpose Driving Scale, PSPDS), revealing that young drivers drove for psychosocial reasons such as for a sense of freedom and to feel independent. Drivers who reported the greatest psychosocial purpose for driving were more likely to be male and to report more risky driving behaviours such as speeding. Drivers who deliberately avoided on-road police presence and reported a prior driving-related offence had significantly greater PSPDS scores, and higher reporting of psychosocial driving purposes was found over time as drivers transitioned from the supervised Learner licence phase to the independent Provisional (intermediate) licence phase. Discussion and conclusions The psychosocial needs met by driving suggest that effective intervention to prevent young driver injury requires further consideration of their driving purpose. Enforcement, education, and engineering efforts which consider the psychosocial purpose of the driving are likely to be more efficacious than those which presently do not. Road safety countermeasures could reduce the young driver’s exposure to risk through such mechanisms as encouraging the use of public transport.

Active learning: A step towards automating medical concept extraction

Objective This paper presents an automatic active learning-based system for the extraction of medical concepts from clinical free-text reports. Specifically, (1) the contribution of active learning in reducing the annotation effort, and (2) the robustness of incremental active learning framework across different selection criteria and datasets is determined. Materials and methods The comparative performance of an active learning framework and a fully supervised approach were investigated to study how active learning reduces the annotation effort while achieving the same effectiveness as a supervised approach. Conditional Random Fields as the supervised method, and least confidence and information density as two selection criteria for active learning framework were used. The effect of incremental learning vs. standard learning on the robustness of the models within the active learning framework with different selection criteria was also investigated. Two clinical datasets were used for evaluation: the i2b2/VA 2010 NLP challenge and the ShARe/CLEF 2013 eHealth Evaluation Lab. Results The annotation effort saved by active learning to achieve the same effectiveness as supervised learning is up to 77%, 57%, and 46% of the total number of sequences, tokens, and concepts, respectively. Compared to the Random sampling baseline, the saving is at least doubled. Discussion Incremental active learning guarantees robustness across all selection criteria and datasets. The reduction of annotation effort is always above random sampling and longest sequence baselines. Conclusion Incremental active learning is a promising approach for building effective and robust medical concept extraction models, while significantly reducing the burden of manual annotation.

Aspect-based opinion mining from product reviews using conditional random fields

Product reviews are the foremost source of information for customers and manufacturers to help them make appropriate purchasing and production decisions. Natural language data is typically very sparse; the most common words are those that do not carry a lot of semantic content, and occurrences of any particular content-bearing word are rare, while co-occurrences of these words are rarer. Mining product aspects, along with corresponding opinions, is essential for Aspect-Based Opinion Mining (ABOM) as a result of the e-commerce revolution. Therefore, the need for automatic mining of reviews has reached a peak. In this work, we deal with ABOM as sequence labelling problem and propose a supervised extraction method to identify product aspects and corresponding opinions. We use Conditional Random Fields (CRFs) to solve the extraction problem and propose a feature function to enhance accuracy. The proposed method is evaluated using two different datasets. We also evaluate the effectiveness of feature function and the optimisation through multiple experiments.

Exercise, Appetite Control, and Body Weight Regulation

Exercise has many health benefits and should be an effective weight loss strategy because it increases energy expenditure. However, the success of exercise in producing and sustaining weight loss is influenced by compensatory changes in energy intake and non-exercise activity, among other factors (see King et al. Obesity 15(6):1373–1383, 2007 for a detailed review). The aim of this chapter is to discuss the evidence describing the relationship between exercise and body weight regulation, with a particular focus on appetite control. Evidence is discussed which demonstrates that weight loss responses to exercise are highly variable between individuals. The mechanisms underlying the relationship between exercise, appetite and energy intake, and hence body weight are also discussed. Some people experience an increase in fasting hunger in response to 12 weeks of supervised exercise. However, this is offset by an increase in meal-related satiety in overweight and obese individuals. It is worth noting that weight loss should not be considered as the only successful outcome of an exercise program. Indeed, exercise, even in the absence of weight loss, is associated with numerous health benefits. Nevertheless, an improved understanding of compensatory responses to exercise is vital so that exercise can be more effectively used in weight management; such an understanding may assist us to devise strategies to sustain greater long-term participation in physical activity.

Improving deep convultional neural networks with unsupervised feature learning

The latest generation of Deep Convolutional Neural Networks (DCNN) have dramatically advanced challenging computer vision tasks, especially in object detection and object classification, achieving state-of-the-art performance in several computer vision tasks including text recognition, sign recognition, face recognition and scene understanding. The depth of these supervised networks has enabled learning deeper and hierarchical representation of features. In parallel, unsupervised deep learning such as Convolutional Deep Belief Network (CDBN) has also achieved state-of-the-art in many computer vision tasks. However, there is very limited research on jointly exploiting the strength of these two approaches. In this paper, we investigate the learning capability of both methods. We compare the output of individual layers and show that many learnt filters and outputs of the corresponding level layer are almost similar for both approaches. Stacking the DCNN on top of unsupervised layers or replacing layers in the DCNN with the corresponding learnt layers in the CDBN can improve the recognition/classification accuracy and training computational expense. We demonstrate the validity of the proposal on ImageNet dataset.