A weighted overlook graph representation of eeg data for absence epilepsy detection
- Authors: Wang, Jialin , Liang, Shen , Wang, Ye , Zhang, Yanchun , Ma, Jiangang
- Date: 2020
- Type: Text , Conference proceedings , Conference paper
- Relation: 20th IEEE International Conference on Data Mining, ICDM 2020 Vol. 2020-November, p. 581-590
- Full Text: false
- Reviewed:
- Description: Absence epilepsy is one of the most common types of epilepsy. The diagnosis of absence epilepsy is among the greatest challenges faced by clinical neurologists due to a lack of easily observable symptoms that are present in conventional epilepsy (e.g. spasm and convulsion), and highly relies on the detection of Spike and Slow Waves (SSWs) in Electroencephalogram (EEG) signals. Recently, graph representations called complex networks have been increasingly applied to characterizing 1D EEG signals. However, existing methods often fail to effectively represent SSWs, struggling to capture the differences between SSW waveforms and their non-SSW counterparts, such as minute differences and distinct shapes. Addressing this issue, in this work, we propose two simple yet effective complex networks, Overlook Graph (OG) and Weighted Overlook Graph (WOG), which have been customized to expressively represent SSWs. Built upon OG and WOG, we then develop a 2D Convolutional Neural Network (2D-CNN) to further learn latent features from the graph representations and accomplish the detection task. Extensive experiments on a real-world absence epilepsy EEG dataset show that the proposed OG/WOG-2D-CNN method can accurately detect SSWs. Additional experiments on the well-known Bonn dataset further show that our method can generalize to the conventional epilepsy seizure detection task with highly competitive performances. © 2020 IEEE. *Please note that there are multiple authors for this article therefore only the name of the first 5 including Federation University Australia affiliate "Jiangang Ma“ is provided in this record**
Human pose based video compression via forward-referencing using deep learning
- Authors: Rajin, S.M. Ataul Karim , Murshed, Manzur , Paul, Manoranjan , Teng, Shyh , Ma, Jiangang
- Date: 2022
- Type: Text , Conference paper
- Relation: 2022 IEEE International Conference on Visual Communications and Image Processing, VCIP 2022, Suzhou, China,13-16 December 2022, 2022 IEEE International Conference on Visual Communications and Image Processing, VCIP 2022
- Full Text: false
- Reviewed:
- Description: To exploit high temporal correlations in video frames of the same scene, the current frame is predicted from the already-encoded reference frames using block-based motion estimation and compensation techniques. While this approach can efficiently exploit the translation motion of the moving objects, it is susceptible to other types of affine motion and object occlusion/deocclusion. Recently, deep learning has been used to model the high-level structure of human pose in specific actions from short videos and then generate virtual frames in future time by predicting the pose using a generative adversarial network (GAN). Therefore, modelling the high-level structure of human pose is able to exploit semantic correlation by predicting human actions and determining its trajectory. Video surveillance applications will benefit as stored 'big' surveillance data can be compressed by estimating human pose trajectories and generating future frames through semantic correlation. This paper explores a new way of video coding by modelling human pose from the already-encoded frames and using the generated frame at the current time as an additional forward-referencing frame. It is expected that the proposed approach can overcome the limitations of the traditional backward-referencing frames by predicting the blocks containing the moving objects with lower residuals. Our experimental results show that the proposed approach can achieve on average up to 2.83 dB PSNR gain and 25.93% bitrate savings for high motion video sequences compared to standard video coding. © 2022 IEEE.
Anomaly detection on health data
- Authors: Samariya, Durgesh , Ma, Jiangang
- Date: 2022
- Type: Text , Conference paper
- Relation: 11th International Conference on Health Information Science, HIS 2022, Virtual, Online, 28- 30 October 2022, Health Information Science, 11th International Conference, HIS 2022, Virtual Event, October 28–30, 2022, Proceedings Vol. 13705 LNCS, p. 34-41
- Full Text: false
- Reviewed:
- Description: The identification of anomalous records in medical data is an important problem with numerous applications such as detecting anomalous reading, anomalous patient health condition, health insurance fraud detection and fault detection in mechanical components. This paper compares the performances of seven state-of-the-art anomaly detection algorithms to do detect anomalies in healthcare data. Our experimental results in six datasets show that the state-of-the-art method of isolation based method iForest has a better performance overall in terms of AUC and runtime. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
Enhancing linear time complexity time series classification with hybrid bag-of-patterns
- Authors: Liang, Shen , Zhang, Yanchun , Ma, Jiangang
- Date: 2020
- Type: Text , Conference paper
- Relation: 25th International Conference on Database Systems for Advanced Applications, DASFAA 2020 Vol. 12112 LNCS, p. 717-735
- Full Text: false
- Reviewed:
- Description: In time series classification, one of the most popular models is Bag-Of-Patterns (BOP). Most BOP methods run in super-linear time. A recent work proposed a linear time BOP model, yet it has limited accuracy. In this work, we present Hybrid Bag-Of-Patterns (HBOP), which can greatly enhance accuracy while maintaining linear complexity. Concretely, we first propose a novel time series discretization method called SLA, which can retain more information than the classic SAX. We use a hybrid of SLA and SAX to expressively and compactly represent subsequences, which is our most important design feature. Moreover, we develop an efficient time series transformation method that is key to achieving linear complexity. We also propose a novel X-means clustering subroutine to handle subclasses. Extensive experiments on over 100 datasets demonstrate the effectiveness and efficiency of our method. © 2020, Springer Nature Switzerland AG.
Mining outlying aspects on healthcare data
- Authors: Samariya, Durgesh , Ma, Jiangang
- Date: 2021
- Type: Text , Conference paper
- Relation: 10th International Conference on Health Information Science, HIS 2021, Melbourne, 25-28 October 2021, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol. 13079 LNCS, p. 160-170
- Full Text: false
- Reviewed:
- Description: Machine learning and artificial intelligence have a wide range of applications in medical domain, such as detecting anomalous reading, anomalous patient health condition, etc. Many algorithms have been developed to solve this problem. However, they fail to answer why those entries are considered as an outlier. This research gap leads to outlying aspect mining problem. The problem of outlying aspect mining aims to discover the set of features (a.k.a subspace) in which the given data point is dramatically different than others. In this paper, we present an interesting application of outlying aspect mining in the medical domain. This paper aims to effectively and efficiently identify outlying aspects using different outlying aspect mining algorithms and evaluate their performance on different real-world healthcare datasets. The experimental results show that the latest isolation-based outlying aspect mining measure, SiNNE, have outstanding performance on this task and have promising results. © 2021, Springer Nature Switzerland AG.
A new effective and efficient measure for outlying aspect mining
- Authors: Samariya, Durgesh , Aryal, Sunil , Ting, Kai , Ma, Jiangang
- Date: 2020
- Type: Text , Conference paper
- Relation: 21st International Conference on Web Information Systems Engineering, WISE 2020, Amsterdam. 20-24 October 2020, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics Vol. 12343 LNCS, p. 463-474
- Full Text: false
- Reviewed:
- Description: Outlying Aspect Mining (OAM) aims to find the subspaces (a.k.a. aspects) in which a given query is an outlier with respect to a given data set. Existing OAM algorithms use traditional distance/density-based outlier scores to rank subspaces. Because these distance/density-based scores depend on the dimensionality of subspaces, they cannot be compared directly between subspaces of different dimensionality. Z-score normalisation has been used to make them comparable. It requires to compute outlier scores of all instances in each subspace. This adds significant computational overhead on top of already expensive density estimation—making OAM algorithms infeasible to run in large and/or high-dimensional datasets. We also discover that Z-score normalisation is inappropriate for OAM in some cases. In this paper, we introduce a new score called Simple Isolation score using Nearest Neighbor Ensemble (SiNNE), which is independent of the dimensionality of subspaces. This enables the scores in subspaces with different dimensionalities to be compared directly without any additional normalisation. Our experimental results revealed that SiNNE produces better or at least the same results as existing scores; and it significantly improves the runtime of an existing OAM algorithm based on beam search. © 2020, Springer Nature Switzerland AG.
Bilateral insider threat detection : harnessing standalone and sequential activities with recurrent neural networks
- Authors: Manoharan, Phavithra , Hong, Wei , Yin, Jiao , Zhang, Yanchun , Ye, Wenjie , Ma, Jiangang
- Date: 2023
- Type: Text , Conference paper
- Relation: 24th International Conference on Web Information Systems Engineering, WISE 2023, Melbourne, 25-27 October 2023, Web Information Systems Engineering – WISE 2023, 24th International Conference, Melbourne, VIC, Australia, October 25–27, 2023, Proceedings Vol. 14306 LNCS, p. 179-188
- Full Text: false
- Reviewed:
- Description: Insider threats involving authorised individuals exploiting their access privileges within an organisation can yield substantial damage compared to external threats. Conventional detection approaches analyse user behaviours from logs, using binary classifiers to distinguish between malicious and non-malicious users. However, existing methods focus solely on standalone or sequential activities. To enhance the detection of malicious insiders, we propose a novel approach: bilateral insider threat detection combining RNNs to incorporate standalone and sequential activities. Initially, we extract behavioural traits from log files representing standalone activities. Subsequently, RNN models capture features of sequential activities. Concatenating these features, we employ binary classification to detect insider threats effectively. Experiments on the CERT 4.2 dataset showcase the approach’s superiority, significantly enhancing insider threat detection using features from both standalone and sequential activities. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
Active model selection for positive unlabeled time series classification
- Authors: Liang, Shen , Zhang, Yanchun , Ma, Jiangang
- Date: 2020
- Type: Text , Conference proceedings , Conference paper
- Relation: 36th IEEE International Conference on Data Engineering, ICDE 2020 Vol. 2020-April, p. 361-372
- Full Text: false
- Reviewed:
- Description: Positive unlabeled time series classification (PUTSC) refers to classifying time series with a set PL of positive labeled examples and a set U of unlabeled ones. Model selection for PUTSC is a largely untouched topic. In this paper, we look into PUTSC model selection, which as far as we know is the first systematic study in this topic. Focusing on the widely adopted self-training one-nearest-neighbor (ST-1NN) paradigm, we propose a model selection framework based on active learning (AL). We present the novel concepts of self-training label propagation, pseudo label calibration principles and ultimately influence to fully exploit the mechanism of ST-1NN. Based on them, we develop an effective model performance evaluation strategy and three AL sampling strategies. Experiments on over 120 datasets and a case study in arrhythmia detection show that our methods can yield top performance in interactive environments, and can achieve near optimal results by querying very limited numbers of labels from the AL oracle. © 2020 IEEE.
- Description: E1
PU-shapelets : Towards pattern-based positive unlabeled classification of time series
- Authors: Liang, Shen , Zhang, Yanchun , Ma, Jiangang
- Date: 2019
- Type: Text , Conference proceedings , Conference paper
- Relation: 24th International Conference on Database Systems for Advanced Applications, DASFAA 2019; Chiang Mai, Thailand; 22nd-25th April 2019; part of the Lecture Notes in Computer Science book series, also part of the Information Systems and Applications, incl. Internet/Web and HCI sub series Vol. 11446 LNCS, p. 87-103
- Full Text:
- Reviewed:
- Description: Real-world time series classification applications often involve positive unlabeled (PU) training data, where there are only a small set PL of positive labeled examples and a large set U of unlabeled ones. Most existing time series PU classification methods utilize all readings in the time series, making them sensitive to non-characteristic readings. Characteristic patterns named shapelets present a promising solution to this problem, yet discovering shapelets under PU settings is not easy. In this paper, we take on the challenging task of shapelet discovery with PU data. We propose a novel pattern ensemble technique utilizing both characteristic and non-characteristic patterns to rank U examples by their possibilities of being positive. We also present a novel stopping criterion to estimate the number of positive examples in U. These enable us to effectively label all U training examples and conduct supervised shapelet discovery. The shapelets are then used to build a one-nearest-neighbor classifier for online classification. Extensive experiments demonstrate the effectiveness of our method.
- Description: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Optimization of parameters of the Kelvin element in vibration analysis
- Authors: Kuznetsov, Alexey , Mammadov, Musa , Hajilarov, Eldar
- Date: 2009
- Type: Text , Conference paper
- Relation: Paper presented at 2009 IEEE International Conference on Industrial Technology, ICIT 2009, Churchill, VIC January 2009
- Full Text:
- Description: In this paper we consider the problem of finding optimal parameters of the Kelvin element in vibration analysis. This problem is based on finding analytical solution of the initial ODE for development of the optimization model. Such technique allows us to compute optimal parameters of Kelvin element.
Clustered memetic algorithm for protein structure prediction
- Authors: Islam, M. D. , Chetty, Madhu
- Date: 2010
- Type: Text , Conference paper
- Relation: Evolutionary Computation (CEC), 2010 IEEE Congress
- Full Text:
- Reviewed:
Oscillations in low-dimensional cyclic differential delay systems
- Authors: Ivanov, Anatoli, , Dzalilov, Zari
- Date: 2018
- Type: Text , Conference paper
- Relation: International conference on Applied Mathematics, Modeling and Computational Science, AMMCS 2017, Waterloo, Canada, August 20–25, 2017 Vol. 259, p. 603-613
- Full Text:
- Reviewed:
- Description: Nonlinear autonomous N-dimensional systems of cyclic differential equations with delays and overall negative feedback are considered. Such systems serve as mathematical models of numerous real world phenomena in physics and laser optics, physiology and mathematical biology, economics and life sciences among others. In the case of lower dimensions and sufficient conditions are derived for the oscillation of all solutions about the unique equilibrium. Open problems and conjectures are discussed for the higher dimensional case and for more convoluted sign feedbacks. © 2018, Springer Nature Switzerland AG.
The implementation of Blockchain framework in MOOCs to support a freedom of learning in Indonesia
- Authors: Febrinanto, Falih , Dafik , Nisviasari, R.
- Date: 2021
- Type: Text , Conference paper
- Relation: 4th International Conference on Combinatorics, Graph Theory, and Network Topology, ICCGANT 2020 Vol. 1836
- Full Text:
- Reviewed:
- Description: A freedom of learning program has been released by the Indonesian Ministry and Culture this year 2020. There are three ways for students to earn their credits, namely take the subject course in face-to-face based class, virtual based class or under Massive Open Online Courses (MOOCs). MOOCs is a model that is developed to help people to learn about certain skills through the online platform, without any limitation in the audience. MOOCs aim to enhance broad collaboration between individuals in creating learning environments that have high scalability and can be accessed by anyone and anywhere. The complexity arises when students undertake a subject course through MOOCs, how to certify the completion of their program in which the certification can be gained easily, and the last how secure the obtained certificate? Blockchain technology can help to improve the quality of MOOCs by providing control of academic records as evidence that someone has completed a learning process on MOOCs. Academic records generated will be stored in one place forever and safely stored in the Blockchain environment. This article will explore how the possible to implement the Blockchain framework in MOOCs to support a freedom of learning in Indonesia. © 2021 Published under licence by IOP Publishing Ltd.
Vibration analysis : Optimization of parameters of the two mass model based on Kelvin elements
- Authors: Kuznetsov, Alexey , Mammadov, Musa , Sultan, Ibrahim , Hajilarov, Eldar
- Date: 2010
- Type: Text , Conference paper
- Relation: Paper presented at 8th IEEE International Conference on Control and Automation, ICCA 2010, Asia Gulf Hotel, Xiamen, China : 9th-11th June 2010 p. 1326-1332
- Full Text:
- Description: In this paper we consider the problem of finding optimal parameters of the two mass model that represents vehicle suspension systems. The analysis of the problem is based on finding analytical solution of the system of coupled Ordinary Differential Equations (ODE). Such a technique allows us to generate optimization problem, where an objective function should be minimized, in accordance with ISO 2631 standard formula of admissible acceleration levels. That ensures maximum comfort for a driver and passenger in a moving vehicle on the considered highways.
- Description: 2003008232
Effects of CaCO3 on kaolin: Fabric, shear strength, and deformation
- Authors: Kim, Sungho , Palomino, Angelica
- Date: 2012
- Type: Text , Conference paper
- Relation: Conference: Experimental Micromechanics for Geomaterials - Joint workshop of the ISSMGE TC101-TC105
- Full Text:
- Reviewed:
Developing a spiritual health and life-orientation measure for secondary school students
- Authors: Fisher, John
- Date: 1999
- Type: Text , Conference paper
- Relation: Paper presented at Research with a regional/rural focus : proceedings of the University of Ballarat inaugural annual conference, Mt. Helen: Victoria 15th October, 1999 p. 57-63
- Full Text:
- Description: The problem posed in this project was the development of an instrument to give a balanced assessment of young people’s spiritual health. Spiritual health is a dynamic state of being, which can be reflected in how well people relate in up to four domains of human existence, namely with themselves; with others; with the environment; and/or with a Transcendent Other. A convenience sample of 850 secondary students in State, Catholic, Christian Community and other independent schools in Ballarat and western suburbs of Melbourne were surveyed during 1999 to determine how important they considered each of the four sets of relationships to be for an ideal state of spiritual health (called Life-Orientation). They also expressed how each area reflected their personal experience most of the time (called Spiritual Health). Extensive factor analysis enabled the original 60-item instrument to be reduced to a reliable, compact 25-item Spiritual Health And Life-Orientation Measure (SHALOM for short). Analysis of variance and t-tests revealed significant variations between students’ views when compared by school type, gender, and year level. SHALOM has advantages over previous instruments in that it is balanced across the four domains of spiritual well-being, is more sensitive, and it compares people’s stated ideal position, with their lived experience, not others’, in determining the quality of relationships which constitute their spiritual well-being.
Characterizations of robust and stable duality for linearly perturbed uncertain optimization problems
- Authors: Dinh, Nguyen , Goberna, Miguel , López, Marco , Volle, Michel
- Date: 2020
- Type: Text , Conference paper
- Relation: Jonathan Borwein Commemorative Conference, JBCC 2017 Vol. 313, p. 43-74
- Relation: http://purl.org/au-research/grants/arc/DP180100602
- Full Text:
- Reviewed:
- Description: We introduce a robust optimization model consisting in a family of perturbation functions giving rise to certain pairs of dual optimization problems in which the dual variable depends on the uncertainty parameter. The interest of our approach is illustrated by some examples, including uncertain conic optimization and infinite optimization via discretization. The main results characterize desirable robust duality relations (as robust zero-duality gap) by formulas involving the epsilon-minima or the epsilon-subdifferentials of the objective function. The two extreme cases, namely, the usual perturbational duality (without uncertainty), and the duality for the supremum of functions (duality parameter vanishing) are analyzed in detail. © Springer Nature Switzerland AG 2020.
On large graphs with given degree and diameter
- Authors: Miller, Mirka , Villavicencio, Guillermo
- Date: 2005
- Type: Text , Conference paper
- Relation: Paper presented at the Sixteenth Australasian Workshop on Combinatorial Algorithms, 18-21 September 2005, Ballarat, Australia, Ballarat, Victoria : 18-12th September, 2005
- Full Text:
- Reviewed:
- Description: E1
- Description: 2003001392
Canonical duality theory and algorithm for solving challenging problems in network optimisation
- Authors: Ruan, Ning , Gao, David
- Date: 2012
- Type: Text , Conference paper
- Relation: 19th International Conference on Neural Information Processing, ICONIP 2012 Vol. 7665 LNCS, p. 702-709
- Full Text:
- Reviewed:
- Description: This paper presents a canonical dual approach for solving a general nonconvex problem in network optimization. Three challenging problems, sensor network location, traveling salesman problem, and scheduling problem are listed to illustrate the applications of the proposed method. It is shown that by the canonical duality, these nonconvex and integer optimization problems are equivalent to unified concave maximization problem over a convex set and hence can be solved efficiently by existing optimization techniques. © 2012 Springer-Verlag.
- Description: 2003010653
The interplay between caste and the emergent accounting profession in India
- Authors: Sidhu, Jasvinder , West, Brian
- Date: 2011
- Type: Text , Conference paper
- Relation: 2011 Afaanz Conference Melbourne 2nd-5th July, 2011 p. 1-23
- Full Text:
- Reviewed:
- Description: This paper examines the interplay between caste and the organised accounting profession in India following the formation of the Institute of Chartered Accountants of India (ICAl) in 1949. The distinctive Indian caste system divides Hindu society into a hierarchy of four "varnas". As well as being traditional markers of social status, these "varnas" have also been identified with particular occupational categories. Using the earliest available membership list of the ICAl (being 1953, just four years after formation), a comprehensive classification of the caste status of the Hindu members of the ICAI was undertaken. This revealed that, relative to the general population, the Brahman upper caste was significantly over-represented in the membership of the ICAl. In this way, the formation of the ICAl predominantly benefitted an already privileged social stratum and contributed to perpetuating traditional caste hierarchies within the context of the modern occupation of professional accounting.
- Description: 2003009025