Fast anomaly detection for streaming data
- Authors: Tan, Swee , Ting, Kaiming , Liu, Fei
- Date: 2011
- Type: Text , Conference paper
- Relation: International Joint Conference on Artificial Intelligence (IJCAI) p. 1511-1516
- Full Text: false
- Reviewed:
- Description: This paper introduces Streaming Half-Space-Trees (HS-Trees), a fast one-class anomaly detector for evolving data streams. It requires only normal data for training and works well when anomalous data are rare. The model features an ensemble of random HS-Trees, and the tree structure is constructed without any data. This makes the method highly efficient because it requires no model restructuring when adapting to evolving data streams. Our analysis shows that Streaming HS-Trees has constant amortised time complexity and constant memory requirement. When compared with a state-of-the-art method, our method performs favourably in terms of detection accuracy and runtime performance. Our experimental results also show that the detection performance of Streaming HS-Trees is not sensitive to its parameter settings.
Feature-subspace aggregating: ensembles for stable and unstable learners
- Authors: Ting, Kaiming , Wells, Jonathan , Tan, Swee , Teng, Shyh , Webb, Geoffrey
- Date: 2011
- Type: Text , Journal article
- Relation: Machine Learning Vol. 82, no. 3 (2011), p. 375-397
- Full Text: false
- Reviewed:
- Description: This paper introduces a new ensemble approach, Feature-Subspace Aggregating (Feating), which builds local models instead of global models. Feating is a generic ensemble approach that can enhance the predictive performance of both stable and unstable learners. In contrast, most existing ensemble approaches can improve the predictive performance of unstable learners only. Our analysis shows that the new approach reduces the execution time to generate a model in an ensemble through an increased level of localisation in Feating. Our empirical evaluation shows that Feating performs significantly better than Boosting, Random Subspace and Bagging in terms of predictive accuracy, when a stable learner SVM is used as the base learner. The speed up achieved by Feating makes feasible SVM ensembles that would otherwise be infeasible for large data sets. When SVM is the preferred base learner, we show that Feating SVM performs better than Boosting decision trees and Random Forests. We further demonstrate that Feating also substantially reduces the error of another stable learner, k-nearest neighbour, and an unstable learner, decision tree.
Improved symmetric-SIFT for Multi-modal image registration
- Authors: Hossain, Md. Tanvir , Lv, Guohua , Teng, Shyh , Lu, Guojun , Lackmann, Martin
- Date: 2011
- Type: Text , Conference paper
- Relation: 2011 International Conference on Digital Image Computing: Techniques and Applications p. 197-202
- Full Text: false
- Reviewed:
- Description: Multi-modal image registration has received significant research attention over the past decade. SymmetricSIFT is a recently proposed local description technique that can be used for registering multi-modal images. It is based on a well-known general image registration technique named Scale Invariant Feature Transform (SIFT). Symmetric-SIFT, however, achieves this invariance to multi-modality at the cost of losing important information. In this paper, we show how this loss may adversely affect the accuracy of registration results. We then propose an improvement to Symmetric-SIFT to overcome the problem. Our experimental results show that the proposed technique can improve the number of true matches by up to 10 times and overall matching accuracy by up to 30%.
Improving SIFT's performance by incorporating appropriate gradient information
- Authors: Lv, Guohua , Hossain, Md. Tanvir , Teng, Shyh , Lu, Guojun , Lackmann, Martin
- Date: 2011
- Type: Text , Conference paper
- Relation: 26th Image and Vision Computing New Zealand Conference (IVCNZ 2011) p. 381 - 386
- Full Text: false
- Reviewed:
- Description: Scale Invariant Feature Transform (SIFT) has been applied in numerous applications especially in the domain of computer vision. In these applications, image information used for building the SIFT descriptor can have a significant impact on its performance. When building orientation histograms for descriptors, a critical step is how to increment the values in the orientation bins. The original scheme for this step in SIFT was improved in [6]. Two different types of gradient information are used for building orientation histograms. The limitations of the two schemes are identified in this paper and we then propose three new schemes which use both types of gradient information in the feature description and matching stages. Our experimental results show that the proposed schemes can achieve better registration performances than the schemes proposed in SIFT and [6].
Information theoretic dynamic Bayesian network approach for reconstructing genetic networks
- Authors: Morshed, Nizamul , Chetty, Madhu
- Date: 2011
- Type: Text , Conference paper
- Relation: Proceedings of the Eleventh IASTED International Conference on Artificial Intelligence and Applications p. 236-243
- Full Text: false
- Reviewed:
- Description: A holistic understanding of genetic interactions, in the post-genomic era, is vital for analysing complex biological systems. In this paper, we present an information theory based novel gene regulatory network inference method using the dynamic Bayesian network (DBN) framework. The proposed approach, with strong theoretical underpinnings, employs mutual information based conditional independence tests to assess the regulatory relationships among genes. The method is flexible, computationally fast and allows a-priori incorporation of biological domain knowledge. We apply it to the analysis of synthetic data as well as Saccharomyces cerevisiae (yeast cell cycle) gene expression data. Performance measures applied to simulation studies show the superior performance of the proposed method
Music classification via the bag-of-features approach
- Authors: Fu, Zhouyu , Lu, Guojun , Ting, Kaiming , Zhang, Dengsheng
- Date: 2011
- Type: Text , Journal article
- Relation: Pattern Recognition Letters Vol. 32, no. 14 (2011), p. 1768-1777
- Full Text: false
- Reviewed:
- Description: A central problem in music information retrieval is audio-based music classification. Current music classification systems follow a frame-based analysis model. A whole song is split into frames, where a feature vector is extracted from each local frame. Each song can then be represented by a set of feature vectors. How to utilize the feature set for global song-level classification is an important problem in music classification. Previous studies have used summary features and probability models which are either overly restrictive in modeling power or numerically too difficult to solve. In this paper, we investigate the bag-of-features approach for music classification which can effectively aggregate the local features for song-level feature representation. Moreover, we have extended the standard bag-of-features approach by proposing a multiple codebook model to exploit the randomness in the generation of codebooks. Experimental results for genre classification and artist identification on benchmark data sets show that the proposed classification system is highly competitive against the standard methods.
Novel local improvement techniques in clustered memetic algorithm for protein structure prediction
- Authors: Islam, Md Kamrul , Chetty, Madhu , Murshed, Manzur
- Date: 2011
- Type: Text , Conference paper
- Relation: IEEE Congress on Evolutionary Computation (IEEE CEC) p. 1003-1011
- Full Text: false
- Reviewed:
- Description: Evolutionary algorithms (EAs) often fail to find the global optimum due to genetic drift. As the protein structure prediction problem is multimodal having several global optima, EAs empowered with combined application of local and global search e.g., memetic algorithms, can be more effective. This paper introduces two novel local improvement techniques for the clustered memetic algorithm to incorporate both problem specific and search-space specific knowledge to find one of the optimum structures of a hydrophobic-polar protein sequence on lattice models. Experimental results show the superiority of the proposed techniques against existing EAs on benchmark sequences.
On dynamic scene geometry for view-invariant action matching
- Authors: Ul-Haq, Anwaar , Gondal, Iqbal , Murshed, Manzur
- Date: 2011
- Type: Text , Conference paper
- Relation: 24th IEEE Conference on Computer Vision and Pattern Recognition (CVPR) p. 3305-3312
- Full Text: false
- Reviewed:
- Description: Variation in viewpoints poses significant challenges to action recognition. One popular way of encoding view-invariant action representation is based on the exploitation of epipolar geometry between different views of the same action. Majority of representative work considers detection of landmark points and their tracking by assuming that motion trajectories for all landmark points on human body are available throughout the course of an action. Unfortunately, due to occlusion and noise, detection and tracking of these landmarks is not always robust. To facilitate it, some of the work assumes that such trajectories are manually marked which is a clear drawback and lacks automation introduced by computer vision. In this paper, we address this problem by proposing view invariant action matching score based on epipolar geometry between actor silhouettes, without tracking and explicit point correspondences. In addition, we explore multi-body epipolar constraint which facilitates to work on original action volumes without any pre-processing. We show that multi-body fundamental matrix captures the geometry of dynamic action scenes and helps devising an action matching score across different views without any prior segmentation of actors. Extensive experimentation on challenging view invariant action datasets shows that our approach not only removes long standing assumptions but also achieves significant improvement in recognition accuracy and retrieval.
On low-rank regularized least squares for scalable nonlinear classification
- Authors: Fu, Zhouyu , Lu, Guojun , Ting, Kaiming , Zhang, Dengsheng
- Date: 2011
- Type: Text , Conference paper
- Relation: International Conference on Neural Information Processing p. 490-499
- Full Text: false
- Reviewed:
- Description: In this paper, we revisited the classical technique of Regularized Least Squares (RLS) for the classification of large-scale nonlinear data. Specifically, we focus on a low-rank formulation of RLS and show that it has linear time complexity in the data size only and does not rely on the number of labels and features for problems with moderate feature dimension. This makes low-rank RLS particularly suitable for classification with large data sets. Moreover, we have proposed a general theorem for the closed-form solutions to the Leave-One-Out Cross Validation (LOOCV) estimation problem in empirical risk minimization which encompasses all types of RLS classifiers as special cases. This eliminates the reliance on cross validation, a computationally expensive process for parameter selection, and greatly accelerate the training process of RLS classifiers. Experimental results on real and synthetic large-scale benchmark data sets have shown that low-rank RLS achieves comparable classification performance while being much more efficient than standard kernel SVM for nonlinear classification. The improvement in efficiency is more evident for data sets with higher dimensions.
Online knowledge validation with prudence analysis in a document management application
- Authors: Dazeley, Richard , Park, Sung Sik , Kang, Byeongho
- Date: 2011
- Type: Text , Journal article
- Relation: Expert Systems with Applications Vol. , no. (2011), p.
- Full Text: false
- Reviewed:
- Description: Prudence analysis (PA) is a relatively new, practical and highly innovative approach to solving the problem of brittleness in knowledge based system (KBS) development. PA is essentially an online validation approach where as each situation or case is presented to the KBS for inferencing the result is simultaneously validated. Therefore, instead of the system simply providing a conclusion, it also provides a warning when the validation fails. Previous studies have shown that a modification to multiple classification ripple-down rules (MCRDR) referred to as rated MCRDR (RM) has been able to achieve strong and flexible results in simulated domains with artificial data sets. This paper presents a study into the effectiveness of RM in an eHealth document monitoring and classification domain using human expertise. Additionally, this paper also investigates what affect PA has when the KBS developer relied entirely on the warnings for maintenance. Results indicate that the system is surprisingly robust even when warning accuracy is allowed to drop quite low. This study of a previously little touched area provides a strong indication of the potential for future knowledge based system development. © 2011 Elsevier Ltd. All rights reserved.
Predictability of moving average rules and nonlinear properties of stock returns: Evidence from the China stock market.
- Authors: Wang, Zhigang , Zeng, Yong , Pan, Heping , Li, Ping
- Date: 2011
- Type: Text , Journal article
- Relation: New mathematics and natural computation Vol. 7, no. 3 (May 2011 2011), p. 267-279
- Full Text: false
- Reviewed:
- Description: This paper investigates t he predictability of moving average rules for the China stock market. We find that buy signals generate higher returns and less volatility, while returns following sell signals are negative and more volatile. Moreover, the bootstrapping results indicate that the asymmetrical patterns of return and volatility between buy and sell signals cannot be explained by four popular linear models of returns, especially the phenomenon of negative sell returns. We then test the nonlinear dynamic process of returns. Although the existing artificial neural network (ANN) model can replicate the negative sell returns, it fails to capture the volatility patterns of buy and sell returns. Furthermore, we introduce the conditional heteroskedasticity structure into the ANN model and find that the revised ANN model cannot only explain the predictability of returns, but can also capture the patterns of buy and sell volatility, which are never achieved by any linear model of returns tested in the related literature. Therefore, we conclude that the moving average trading rules can pick up some of the hidden nonlinear patterns in the dynamic process of stock returns, which may be the reason why they can be used to predict price changes.
Preface
- Authors: Pan, Heping , Hayward, Serge
- Date: 2011
- Type: Text , Journal article
- Relation: New Mathematics and Natural Computation Vol. 7, no. 2 (2011), p. 187-196
- Full Text: false
- Reviewed:
Resonant frequency band estimation using adaptive wavelet decomposition level selection
- Authors: Yaqub, Muhammad , Gondal, Iqbal , Kamruzzaman, Joarder
- Date: 2011
- Type: Text , Conference paper
- Relation: 2011 IEEE International Conference on Mechatronics and Automation (ICMA) p. 376-381
- Full Text: false
- Reviewed:
- Description: The vibrations induced by machine faults help in diagnosis and prognosis of the machine. It is crucial for the fault diagnostic system to extract resonant frequency band which carries useful information about the defect frequencies and contains maximum signal to noise ratio. The spectral orientation of the resonant frequency band varies with the variation in machine dynamics. The existing techniques which employ wavelet transformation to exploit the signal energy distribution among different frequency sub-bands, are based on fixed decomposition level and do not optimize the wavelet parameters according to varying machine dynamics. The proposed study develops a novel technique: Adaptive Wavelet Decomposition and Resonance Frequency Estimation (AWRE) which estimates the positioning of the resonant frequency band based on adaptive selection of the wavelet decomposition levels. The results for the simulated as well as actual vibration data demonstrate that the proposed technique estimates the bandwidth of the resonant frequency band quite effectively.
Severity invariant machine fault diagnosis
- Authors: Yaqub, Muhammad , Gondal, Iqbal , Kamruzzaman, Joarder
- Date: 2011
- Type: Text , Conference paper
- Relation: 6th IEEE Conference on Industrial Electronics and Applications p. 21-26
- Full Text: false
- Reviewed:
- Description: Vibration signals used for abnormality detection in machine health monitoring (MHM) suffer from significant variation with fault severity. This variation causes overlap among the features belonging to different types of faults resulting in severe degradation of fault detection accuracy. This paper identifies a new problem due to severity variant features and proposes a novel adaptive training set and feature selection (ATSFS) scheme based upon the orientation of the test data. In order to build ATSFS and validate its performance, training and testing data are obtained from different severity levels. To capture the non-stationary behavior of vibration signal, robust tools such as wavelet transform (WT) for time-frequency analysis are employed. Simulation studies show that ATSFS attains high classification accuracy even if training and testing data belong to different severity levels.
Simultaneous learning of instantaneous and time-delayed genetic interactions using novel information theoretic scoring technique
- Authors: Morshed, Nizamul , Chetty, Madhu , Xuan Vinh, Nguyen
- Date: 2011
- Type: Text , Conference paper
- Relation: International Conference on Neural Information Processing p. 248-257
- Full Text: false
- Reviewed:
- Description: Understanding gene interactions is a fundamental question in systems biology. Currently, modeling of gene regulations assumes that genes interact either instantaneously or with time delay. In this paper, we introduce a framework based on the Bayesian Network (BN) formalism that can represent both instantaneous and time-delayed interactions between genes simultaneously. Also, a novel scoring metric having firm mathematical underpinnings is then proposed that, unlike other recent methods, can score both interactions concurrently and takes into account the biological fact that multiple regulators may regulate a gene jointly, rather than in an isolated pair-wise manner. Further, a gene regulatory network inference method employing evolutionary search that makes use of the framework and the scoring metric is also presented. Experiments carried out using synthetic data as well as the well known Saccharomyces cerevisiae gene expression data show the effectiveness of our approach.
Texture classification using multimodal invariant local binary pattern
- Authors: Sadat, Rafi , Teng, Shyh , Lu, Guojun , Hasan, Sheikh
- Date: 2011
- Type: Text , Conference paper
- Relation: IEEE Workshop on Applications of Computer Vision (WACV) p. 315-320
- Full Text: false
- Reviewed:
- Description: As texture information among pixels can be effectively represented using Local binary patterns (LBPs), image descriptors built using LBPs or its variants have been frequently used for various image analysis applications, e.g. medical image and texture image classification and retrieval. However, neither LBP nor any of its existing variants can be used to build descriptors for classifying multimodal images effectively. This is because the same object when captured in different modalities may result in opposite pixel intensity in some corresponding parts of the images, which in turn will cause their descriptors to be very different. To solve this problem, we propose a novel modality invariant texture descriptor which is built by modifying the standard procedure for building LBP. In this paper, we explain how the proposed descriptor can be built efficiently. We also demonstrate empirically that compared to all the state of the art LBP-based descriptors, the proposed descriptor achieves better accuracy for classifying multimodal images
A comparative study of practical stochastic clustering method with traditional methods
- Authors: Tan, Swee , Ting, Kaiming , Teng, Shyh
- Date: 2010
- Type: Text , Conference paper
- Relation: Proceedings of the 23rd Australasian Joint Conference on Artificial Intelligence p. 112-121
- Full Text: false
- Reviewed:
- Description: In many real-world clustering problems, there usually exist little information about the clusters underlying a certain dataset. For example, the number of clusters hidden in many datasets is usually not known a priori. This is an issue because many traditional clustering methods require such information as input. This paper examines a practical stochastic clustering method (PSCM) that has the ability to find clusters in datasets without requiring users to specify the centroids or the number of clusters. By comparing with traditional methods (k-means, self-organising map and hierarchical clustering methods), the performance of PSCM is found to be robust against overlapping clusters and clusters with uneven sizes. The proposed method also scales well with datasets having varying number of clusters and dimensions. Finally, our experimental results on real-world data confirm that the proposed method performs competitively against the traditional clustering methods in terms of clustering accuracy and efficiency.
A comparative study on contour-based corner detectors
- Authors: Awrangjeb, Mohammad , Lu, Guojun , Fraser, Clive
- Date: 2010
- Type: Text , Conference paper
- Relation: Digital Image Computing: Techniques and Applications (DICTA), 2010 International Conference
- Full Text: false
- Reviewed:
- Description: Contour-based corner detectors directly or indirectly estimate a significance measure (e.g. curvature) on the points of a planar curve and select the curvature extrema points as corners. While an extensive number of contour-based corner detectors have been proposed over the last four decades, there is no comparative study of recently proposed promising detectors. This paper is an attempt to fill this gap. We present the general frame-work of the contour-based corner detection technique and discuss two major issues - curve smoothing and curvature estimation, which have major impacts on the corner detection performance. A number of promising detectors are compared using an automatic evaluation system on a common large dataset. It is observed that while the detectors using indirect curvature estimation techniques are more robust, the detectors using direct curvature estimation techniques are faster.
A novel color image fusion QoS measure for multi-sensor night vision applications
- Authors: Anwaar, Ul-Haq , Gondal, Iqbal , Murshed, Manzur
- Date: 2010
- Type: Text , Conference proceedings
- Full Text: false
- Description: Color image fusion of visible and infra-red imagery can play an important role in multi-sensor night vision systems that are an integral part of modern warfare. Image fusion minimizes the amount of required bandwidth by transmitting the fused image rather than multiple sensor images. Color image fusion can be achieved by combining inputs from original colored sensors or by employing pseudo colorization and color transfer to grayscale images. Various quality measures have been proposed for multi-sensor grayscale image fusion techniques; but no appropriate quality measure has been devised for the quality evaluation of multi-sensor color image fusion. In this paper, we propose a novel color image fusion quality measure, Color Fusion Objective Index (CFOI) based on colorfulness, gradient similarity and mutual information techniques. Experimental results show the effectiveness of CFOI to evaluate the color and salient feature extraction introduced by color fusion techniques into the final fused imagery as well as its consistency with subjective evaluation.
A numerical control algorithm for navigation of an operator-driven snake-like robot with 4WD-4WS segments
- Authors: Percy, Andrew , Spark, Ian
- Date: 2010
- Type: Text , Journal article
- Relation: Robotica Vol. 29, no. 3 (2010), p. 471-482
- Full Text:
- Reviewed:
- Description: This paper presents a new algorithm for the control of a snake-like robot with passive joints and active wheels. Each segment has four autonomously driven and steered wheels. The algorithm approximates the ideal solution in which all wheels on a segment have the same centre of curvature with wheel speeds, providing cooperative redundancy. Each hitch point joining segments traverses the same path, which is determined by an operator, prescribing the path curvature and front hitch speed. The numerical algorithm developed in this paper is simulation tested against a previously derived analytical solution for a predetermined path. Further simulations are carried out to show the effects of changing curvature and front hitch speed on hitch path, wheel angles and wheel speeds for a one, two and three segment robot.