Search Google Appliance

Publications

| 1975 | 1976 | 1977 | 1978 | 1979 | 1980 | 1981 | 1982 | 1983 | 1984 | 1985 | 1986 | 1987 | 1988 | 1989 | 1990 | 1991 | 1992 | 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017

Chapters

Adams, J., Sibbritt, D., Broom, A., Kroll, T., Prior, J., Dunston, R., Leung, B., Davidson, P. & Andrews, G. 2017, 'Traditional, complementary and integrative medicine as self-care in chronic illness' in Adams, J. & et al (eds), Public Health and Health Services Research in Traditional, Complementary and Integrative Medicine: International Perspectives, Imperial College Press, London.

Stoianoff, N.P., Cahill, A. & Wright, E.A. 2017, 'Indigenous knowledge: what are the issues?' in Stoianoff, N.P. (ed), Indigenous Knowledge Forum: Comparative Systems for Recognising and Protecting Indigenous Knowledge and Culture, LexisNexis, pp. 11-37.

Wright, E.A., Cahill, A. & Stoianoff, N.P. 2017, 'Australia and Indigenous traditional knowledge' in Stoianoff, N.P. (ed), Indigenous Knowledge Forum: Comparative Systems for Recognising and Protecting Indigenous Knowledge and Culture, LexisNexis, pp. 39-68.

Journal articles

Chai, R., Naik, G.R., Ling, S.H. & Nguyen, H.T. 2017, 'Hybrid brain–computer interface for biomedical cyber-physical system application using wireless embedded EEG systems', BioMedical Engineering OnLine, vol. 16, no. 5, pp. 1-23.
View/Download from: UTS OPUS or Publisher's site

Chiarella, C., He, X.Z., Shi, L. & Wei, L. 2017, 'A behavioural model of investor sentiment in limit order markets', Quantitative Finance, vol. 17, no. 1, pp. 71-86.
View/Download from: Publisher's site
View description>>

© 2016 Informa UK Limited, trading as Taylor & Francis GroupBy incorporating behavioural sentiment in a model of a limit order market, we show that behavioural sentiment not only helps to replicate most of the stylized facts in limit order markets simultaneously, but it also plays a unique role in explaining those stylized facts that cannot be explained by noise trading, such as fat tails in the return distribution, long memory in the trading volume, an increasing and non-linear relationship between trade imbalance and mid-price returns, as well as the diagonal effect, or event clustering, in order submission types. The results show that behavioural sentiment is an important driving force behind many of the well-documented stylized facts in limit order markets.

Du, B., Wang, Z., Zhang, L., Zhang, L., Liu, W., Shen, J. & Tao, D. 2017, 'Exploring Representativeness and Informativeness for Active Learning', IEEE Transactions on Cybernetics, vol. 47, no. 1, pp. 14-26.
View/Download from: Publisher's site
View description>>

How can we find a general way to choose the most suitable samples for training a classifier? Even with very limited prior information? Active learning, which can be regarded as an iterative optimization procedure, plays a key role to construct a refined training set to improve the classification performance in a variety of applications, such as text analysis, image recognition, social network modeling, etc. Although combining representativeness and informativeness of samples has been proven promising for active sampling, state-of-the-art methods perform well under certain data structures. Then can we find a way to fuse the two active sampling criteria without any assumption on data? This paper proposes a general active learning framework that effectively fuses the two criteria. Inspired by a two-sample discrepancy problem, triple measures are elaborately designed to guarantee that the query samples not only possess the representativeness of the unlabeled data but also reveal the diversity of the labeled data. Any appropriate similarity measure can be employed to construct the triple measures. Meanwhile, an uncertain measure is leveraged to generate the informativeness criterion, which can be carried out in different ways. Rooted in this framework, a practical active learning algorithm is proposed, which exploits a radial basis function together with the estimated probabilities to construct the triple measures and a modified best-versus-second-best strategy to construct the uncertain measure, respectively. Experimental results on benchmark datasets demonstrate that our algorithm consistently achieves superior performance over the state-of-the-art active learning algorithms.

Du, B., Zhang, M., Zhang, L., Hu, R. & Tao, D. 2017, 'PLTD: Patch-Based Low-Rank Tensor Decomposition for Hyperspectral Images', IEEE Transactions on Multimedia, vol. 19, no. 1, pp. 67-79.
View/Download from: Publisher's site
View description>>

© 1999-2012 IEEE.Recent years has witnessed growing interest in hyperspectral image (HSI) processing. In practice, however, HSIs always suffer from huge data size and mass of redundant information, which hinder their application in many cases. HSI compression is a straightforward way of relieving these problems. However, most of the conventional image encoding algorithms mainly focus on the spatial dimensions, and they need not consider the redundancy in the spectral dimension. In this paper, we propose a novel HSI compression and reconstruction algorithm via patch-based low-rank tensor decomposition (PLTD). Instead of processing the HSI separately by spectral channel or by pixel, we represent each local patch of the HSI as a third-order tensor. Then, the similar tensor patches are grouped by clustering to form a fourth-order tensor per cluster. Since the grouped tensor is assumed to be redundant, each cluster can be approximately decomposed to a coefficient tensor and three dictionary matrices, which leads to a low-rank tensor representation of both the spatial and spectral modes. The reconstructed HSI can then be simply obtained by the product of the coefficient tensor and dictionary matrices per cluster. In this way, the proposed PLTD algorithm simultaneously removes the redundancy in both the spatial and spectral domains in a unified framework. The extensive experimental results on various public HSI datasets demonstrate that the proposed method outperforms the traditional image compression approaches and other tensor-based methods.

He, X. & Shi, L. 2017, 'Index Portfolio and Welfare Analysis Under Heterogeneous Beliefs', Journal of Banking and Finance, vol. 75, pp. 64-79.
View/Download from: Publisher's site

He, X.Z., Lütkebohmert, E. & Xiao, Y. 2017, 'Rollover Risk and Credit Risk under Time-varying Margin', Quantitative Finance, vol. 17, no. 3, pp. 455-469.

Ho-Le, T.P., Center, J.R., Eisman, J.A., Nguyen, H.T. & Nguyen, T.V. 2017, 'Prediction of Bone Mineral Density and Fragility Fracture by Genetic Profiling.', Journal of Bone and Mineral Research, vol. 32, no. 2, pp. 285-293.
View/Download from: UTS OPUS or Publisher's site
View description>>

Although the susceptibility to fracture is partly determined by genetic factors, the contribution of newly discovered genetic variants to fracture prediction is still unclear. This study sought to define the predictive value of a genetic profiling for fracture prediction.Sixty-two bone mineral density (BMD)-associated single-nucleotide polymorphism (SNP) were genotyped in 557 men and 902 women who had participated in the Dubbo Osteoporosis Epidemiology Study. The incidence of fragility fracture was ascertained from X-ray reports between 1990 and 2015. Femoral neck BMD was measured by dual-energy X-ray absorptiometry. A weighted polygenic risk score (GRS) was created as a function of the number of risk alleles and their BMD-associated regression coefficients for each SNP. The association between GRS and fracture risk was assessed by the Cox's proportional hazards model.Individuals with greater GRS had lower femoral neck BMD (P < 0.01), but the variation in GRS accounted for less than 2% of total variance in BMD. Each unit increase in GRS was associated with a hazard ratio of 1.20 (95%CI, 1.04-1.38) for fracture, and this association was independent of age, prior fracture, fall, and in a subset of 33 SNPs, independent of femoral neck BMD. The significant association between GRS and fracture was observed for the vertebral and wrist fractures, but not for hip fracture. The area under the receiver operating characteristic (ROC) curve for the model with GRS and clinical risk factors was 0.71 (95% CI, 0.68-0.74). With GRS, the correct reclassification of fracture vs non-fracture ranged from 12% for hip fracture to 23% for wrist fracture.A genetic profiling of BMD-associated genetic variants could improve the accuracy of fracture prediction over and above that of clinical risk factors alone, and help stratify individuals by fracture status. This article is protected by copyright. All rights reserved.

Huang, T., Huang, M.L., Nguyen, Q., Zhao, L., Huang, W. & Chen, J. 2017, 'A Space-Filling Multidimensional Visualization (SFMDVis) for Exploratory Data Analysis', Information Sciences, vol. 390, pp. 32-53.
View/Download from: UTS OPUS or Publisher's site
View description>>

The space-filling visualization model was first invented by Ben Shneiderman [28] for maximizing the utilization of display space in relational data (or graph) visualization, especially for tree visualization. It uses the concept of Enclosure which dismisses the “edges” in the graphic representation that are all too frequently used in traditional node-link based graph visualizations. Therefore, the major issue in graph visualization which is the edge crossing can be naturally solved through the adoption of a space filling approach. However in the past, the space-filling concept has not attracted much attention from researchers in the field of multidimensional visualization. Although the problem of ‘edge crossing’ has also occurred among polylines which are used as the basic visual elements in the parallel coordinates visualization, it is problematic if those ‘edge crossings’ among polylines are not evenly distributed on the display plate as visual clutter will occur. This problem could significantly reduce the human readability in terms of reviewing a particular region of the visualization. In this study, we propose a new Space-Filling Multidimensional Data Visualization (SFMDVis) that for the first-time introduces a space-filling approach into multidimensional data visualization. The main contributions are: (1) achieving the maximization of space utilization in multidimensional visualization (i.e. 100% of the display area is fully used), (2) eliminating visual clutter in SFMDVis through the use of the non-classic geometric primitive and (3) improving the quality of visualization for the visual perception of linear correlations among different variables as well as recognizing data patterns. To evaluate the quality of SFMDVis, we have conducted a usability study to measure the performance of SFMDVis in comparison with parallel coordinates and a scatterplot matrix for finding linear correlations and data patterns. The evaluation results have suggested that the acc...

Liu, T., Tao, D., Song, M. & Maybank, S. 2017, 'Algorithm-Dependent Generalization Bounds for Multi-Task Learning.', IEEE Trans Pattern Anal Mach Intell, vol. 39, no. 2, pp. 227-241.
View/Download from: Publisher's site
View description>>

Often, tasks are collected for multi-task learning (MTL) because they share similar feature structures. Based on this observation, in this paper, we present novel algorithm-dependent generalization bounds for MTL by exploiting the notion of algorithmic stability. We focus on the performance of one particular task and the average performance over multiple tasks by analyzing the generalization ability of a common parameter that is shared in MTL. When focusing on one particular task, with the help of a mild assumption on the feature structures, we interpret the function of the other tasks as a regularizer that produces a specific inductive bias. The algorithm for learning the common parameter, as well as the predictor, is thereby uniformly stable with respect to the domain of the particular task and has a generalization bound with a fast convergence rate of order O(1=n), where n is the sample size of the particular task. When focusing on the average performance over multiple tasks, we prove that a similar inductive bias exists under certain conditions on the feature structures. Thus, the corresponding algorithm for learning the common parameter is also uniformly stable with respect to the domains of the multiple tasks, and its generalization bound is of the order O(1=T ), where T is the number of tasks. These theoretical analyses naturally show that the similarity of feature structures in MTL will lead to specific regularizations for predicting, which enables the learning algorithms to generalize fast and correctly from a few examples.

Qiao, M., Liu, L., Yu, J., Xu, C. & Tao, D. 2017, 'Diversified dictionaries for multi-instance learning', Pattern Recognition, vol. 64, pp. 407-416.
View/Download from: Publisher's site
View description>>

© 2016 Elsevier LtdMultiple-instance learning (MIL) has been a popular topic in the study of pattern recognition for years due to its usefulness for such tasks as drug activity prediction and image/text classification. In a typical MIL setting, a bag contains a bag-level label and more than one instance/pattern. How to bridge instance-level representations to bag-level labels is a key step to achieve satisfactory classification accuracy results. In this paper, we present a supervised learning method, diversified dictionaries MIL, to address this problem. Our approach, on the one hand, exploits bag-level label information for training class-specific dictionaries. On the other hand, it introduces a diversity regularizer into the class-specific dictionaries to avoid ambiguity between them. To the best of our knowledge, this is the first time that the diversity prior is introduced to solve the MIL problems. Experiments conducted on several benchmark (drug activity and image/text annotation) datasets show that the proposed method compares favorably to state-of-the-art methods.

Xiong, W., Zhang, L., Du, B. & Tao, D. 2017, 'Combining local and global: Rich and robust feature pooling for visual recognition', Pattern Recognition, vol. 62, pp. 225-235.
View/Download from: Publisher's site
View description>>

© 2016 Elsevier LtdThe human visual system proves expert in discovering patterns in both global and local feature space. Can we design a similar way for unsupervised feature learning? In this paper, we propose a novel spatial pooling method within an unsupervised feature learning framework, named Rich and Robust Feature Pooling (R2FP), to better extract rich and robust representation from sparse feature maps learned from the raw data. Both local and global pooling strategies are further considered to instantiate such a method. The former selects the most representative features in the sub-region and summarizes the joint distribution of the selected features, while the latter is utilized to extract multiple resolutions of features and fuse the features with a feature balance kernel for rich representation. Extensive experiments on several image recognition tasks demonstrate the superiority of the proposed method.

Zhang, T., Jia, W., Yang, B., Yang, J., He, X. & Zheng, Z. 2017, 'MoWLD: a robust motion image descriptor for violence detection', Multimedia Tools and Applications, vol. 76, no. 1, pp. 1419-1438.
View/Download from: UTS OPUS or Publisher's site
View description>>

© 2015 Springer Science+Business Media New York Automatic violence detection from video is a hot topic for many video surveillance applications. However, there has been little success in designing an algorithm that can detect violence in surveillance videos with high performance. Existing methods typically apply the Bag-of-Words (BoW) model on local spatiotemporal descriptors. However, traditional spatiotemporal features are not discriminative enough, and also the BoW model roughly assigns each feature vector to only one visual word and therefore ignores the spatial relationships among the features. To tackle these problems, in this paper we propose a novel Motion Weber Local Descriptor (MoWLD) in the spirit of the well-known WLD and make it a powerful and robust descriptor for motion images. We extend the WLD spatial descriptions by adding a temporal component to the appearance descriptor, which implicitly captures local motion information as well as low-level image appear information. To eliminate redundant and irrelevant features, the non-parametric Kernel Density Estimation (KDE) is employed on the MoWLD descriptor. In order to obtain more discriminative features, we adopt the sparse coding and max pooling scheme to further process the selected MoWLDs. Experimental results on three benchmark datasets have demonstrated the superiority of the proposed approach over the state-of-the-arts.