Asymmetric clustering using the alpha–beta divergence

We propose the use of an asymmetric dissimilarity measure in centroid-based clustering. The dissimilarity employed is the Alpha–Beta divergence (AB-divergence), which can be asymmetrized using its parameters. We compute the degree of asymmetry of the AB-divergence on the basis of the within-cluster variances. In this way, the proposed approach is able to flexibly model even clusters with significantly different variances. Consequently, this method overcomes one of the major drawbacks of the standard symmetric centroid-based clustering.     

Dempster–Shafer clustering using Potts spin mean field theory

 In this article we investigate a problem within Dempster–Shafer theory where 2 ^q −1 pieces of evidence are clustered into q clusters by minimizing a metaconflict function, or equivalently, by minimizing the sum of weight of conflict over all clusters. Previously one of us developed a method based on a Hopfield and Tank model. However, for very large problems we need a method with lower computational complexity. We demonstrate that the weight of conflict of evidence can, as an approximation, be linearized and mapped to an antiferromagnetic Potts spin model. This facilitates efficient numerical solution, even for large problem sizes. Optimal or nearly optimal solutions are found for Dempster–Shafer clustering benchmark tests with a time complexity of approximately O(N ²log² N). Furthermore, an isomorphism between the antiferromagnetic Potts spin model and a graph optimization problem is shown. The graph model has dynamic variables living on the links, which have a priori probabilities that are directly related to the pairwise conflict between pieces of evidence. Hence, the relations between three different models are shown.     

Comparing dynamics of fluency and inter-limb coordination in climbing activities using multi-scale Jensen–Shannon embedding and clustering

This paper reports the results of two studies carried out in a controlled environment aiming to understand relationships between movement patterns of coordination that emerge during climbing and performance outcomes. It involves a recent method of nonlinear dimensionality reduction, multi-scale Jensen–Shannon neighbor embedding (Lee et al., 2015), which has been applied to recordings of movement sensors in order to visualize coordination patterns adapted by climbers. Initial clustering at the climb scale provides details linking behavioral patterns with climbing fluency/smoothness (i.e., the performance outcome). Further clustering on shorter time intervals, where individual actions within a climb are analyzed, enables more detailed exploratory data analysis of behavior. Results suggest that the nature of individual learning curves (the global, trial-to-trial performance) corresponded to certain behavioral patterns (the within trial motor behavior). We highlight and discuss three distinctive learning curves and their corresponding relationship to behavioral pattern emergence, namely: no improvement and a lack of new motor behavior emergence; sudden improvement and the emergence of new motor behaviors; and gradual improvement and a lack of new motor behavior emergence.     

An unsupervised approach for co-channel speech separation using Hilbert–Huang transform and Fuzzy C-Means clustering

In this paper we discuss an unsupervised approach for co-channel speech separation where two speakers are speaking simultaneously over same channel. We propose a two stage separation process where the initial stage is based on empirical mode decomposition (EMD) and Hilbert transform generally known as Hilbert–Huang transform. EMD decomposes the mixed signal into oscillatory functions known as intrinsic mode functions. Hilbert transform is applied to find the instantaneous amplitudes and Fuzzy C-Means clustering is applied to group the speakers at initial stage. In second stage of separation speaker groups are transformed into time–frequency domain using short time Fourier transform (STFT). Time–frequency ratio’s are computed by dividing the STFT matrix of mixed speech signal and STFT matrix of stage1 recovered speech signals. Histogram of the ratios obtained can be used to estimate the ideal binary mask for each speaker. These masks are applied to the speech mixture and the underlying speakers are estimated. Masks are estimated from the speech mixture and helps in imputing the missing values after stage1 grouping of speakers. Results obtained show significant improvement in objective measures over other existing single-channel speech separation methods.     

Feature clustering and feature discretization assisting gene selection for molecular classification using fuzzy c-means and expectation–maximization algorithm

The Journal of Supercomputing - In this paper, a novel gene selection benefiting from feature clustering and feature discretization is developed. In large numbers of genes, unsupervised fuzzy...     

A cloud-based recommendation service using principle component analysis–scale-invariant feature transform algorithm

Cloud computing delivers resources such as software, data, storage and servers over the Internet; its adaptable infrastructure facilitates on-demand access of computational resources. There are many benefits of cloud computing such as being scalable, paying only for consumption, improving accessibility, limiting investment costs and being environmentally friendly. Thus, many organizations have already started applying this technology to improve organizational efficiency. In this study, we developed a cloud-based book recommendation service that uses a principle component analysis–scale-invariant feature transform (PCA-SIFT) feature detector algorithm to recommend book(s) based on a user-uploaded image of a book or collection of books. The high dimensionality of the image is reduced with the help of a principle component analysis (PCA) pre-processing technique. When the mobile application user takes a picture of a book or a collection of books, the system recognizes the image(s) and recommends similar books. The computational task is performed via the cloud infrastructure. Experimental results show the PCA-SIFT-based cloud recommendation service is promising; additionally, the application responds faster when the pre-processing technique is integrated. The proposed generic cloud-based recommendation system is flexible and highly adaptable to new environments.

     

Automated exploration of datapath and unrolling factor during power–performance tradeoff in architectural synthesis using multi-dimensional PSO algorithm

A novel algorithm for automated simultaneous exploration of datapath and Unrolling Factor (UF) during power–performance tradeoff in High Level Synthesis (HLS) using multi-dimensional particle swarm optimization (PSO) (termed as ‘M-PSO’) for control and data flow graphs (CDFGs) is presented. The major contributions of the proposed algorithm are as follows: (a) simultaneous exploration of datapath and loop UF through an integrated multi-dimensional particle encoding process using swarm intelligence; (b) an estimation model for computation of execution delay of a loop unrolled CDFG (based on a resource configuration visited) without requiring to tediously unroll the entire CDFG for the specified loop value in most cases; (c) balancing the tradeoff between power–performance metrics as well as control states and execution delay during loop unrolling; (d) sensitivity analysis of PSO parameter such as swarm size on the impact of exploration time and Quality of Results (QoR) of the proposed design space exploration (DSE) process. This analysis presented would assist the designer in pre-tuning the PSO parameters to an optimum value for achieving efficient exploration results within a quick runtime; (e) analysis of design metrics such as power, execution time and number of control steps of the global best particle found in every iteration with respect to increase/decrease in unrolling factor.The proposed approach when tested on a variety of data flow graphs (DFGs) and CDFGs indicated an average improvement in QoR of >28% and reduction in runtime of >94% compared to recent works.     

Similarity assessment for scientific workflow clustering and recommendation基于相似性评估策略的科学工作流聚类与推荐

This article proposes to identify and recommend scientific workflows for reuse and repurposing. Specifically, a scientific workflow is represented as a layer hierarchy that specifies the hierarchical relations between this workflow, its sub-workflows, and activities. Semantic similarity is calculated between layer hierarchies of workflows. A graph-skeleton based clustering technique is adopted for grouping layer hierarchies into clusters. Barycenters in each cluster are identified, which serve as core workflows in this cluster, for facilitating the cluster identification and workflow ranking and recommendation with respect to the requirement of scientists.     

Unbalanced and Partial $$L_1$$ Monge–Kantorovich Problem: A Scalable Parallel First-Order Method

We propose a new algorithm to solve the unbalanced and partial \(L_1\)-Monge–Kantorovich problems. The proposed method is a first-order primal-dual method that is scalable and parallel. The method’s iterations are conceptually simple, computationally cheap, and easy to parallelize. We provide several numerical examples solved on a CUDA GPU, which demonstrate the method’s practical effectiveness.     

Age and gender classification using brain–computer interface

Neural Computing and Applications - With the development of Internet of things (IOT), it is now possible to connect various heterogeneous devices together using Internet. The devices are able to...     

Selective particle and cell clustering at air–liquid interfaces within ultrasonic microfluidic systems

In this study, we demonstrate particle and cell clustering in distinct patterns on the free surface of microfluidic volumes. Employing ultrasonic actuation, submersed microparticles are forced to two principal positions: nodal lines (pressure minima) of a standing wave within the liquid bulk, and distinct locations on the air–liquid interface (free surface); the latter of which has not been previously demonstrated using ultrasonic standing waves. As such, we unravel the fundamental mechanisms behind such patterns, showing that the contribution of fluid particle velocity variations on the free surface (acoustic radiation force) results in patterned particle clustering. In addition, by varying the size and density of the microparticles (3.5–31 μm polystyrene and 1–5 μm silica), acoustic streaming is found to increase the tendency for a smaller and lighter particle to cluster at the air–liquid interface. This selectivity is exploited for the isolation of multiple microparticle and cell types on the free surface from their nodally aligned counterparts. Free surface clustering is demonstrated in both an open microfluidic chamber and a sessile droplet, as well as using a range of biological species Escherichia coli, blood cells, Ragweed pollen and Paper Mulberry pollen). The ability to selectively cluster submersed microparticles and cells in distinct patterns on the free surface showcases the excellent suitability of this method to lab-on-a-chip systems.     

Pulmonary crackle detection using time–frequency and time–scale analysis

Pulmonary crackles are used as indicators for the diagnosis of different pulmonary disorders in auscultation. Crackles are very common adventitious transient sounds. From the characteristics of crackles such as timing and number of occurrences, the type and the severity of the pulmonary diseases may be assessed. In this study, a method is proposed for crackle detection. In this method, various feature sets are extracted using time–frequency and time–scale analysis from pulmonary signals. In order to understand the effect of using different window and wavelet types in time–frequency and time–scale analysis in detecting crackles, different windows and wavelets are tested such as Gaussian, Blackman, Hanning, Hamming, Bartlett, Triangular and Rectangular windows for time–frequency analysis and Morlet, Mexican Hat and Paul wavelets for time–scale analysis. The extracted feature sets, both individually and as an ensemble of networks, are fed into three different machine learning algorithms: Support Vector Machines, k-Nearest Neighbor and Multilayer Perceptron. Moreover, in order to improve the success of the model, prior to the time–frequency/scale analysis, frequency bands containing no-crackle information are removed using dual-tree complex wavelet transform, which is a shift invariant transform with limited redundancy compared to the conventional discrete wavelet transform. The comparative results of individual feature sets and ensemble of sets, which are extracted using different window and wavelet types, for both pre-processed and non-pre-processed data with different machine learning algorithms, are extensively evaluated and compared.     

Feature extraction using median–mean and feature line embedding

In this article, we propose a feature extraction method based on median–mean and feature line embedding (MMFLE) for the classification of hyperspectral images. In MMFLE, we maximize the class separability using discriminant analysis. Moreover, we remove the negative effect of outliers on the class mean using the median–mean line (MML) measurement and virtually enlarge the training set using the feature line (FL) distance metric. The experimental results on Indian Pines and University of Pavia data sets show the better performance of MMFLE compared to nearest feature line embedding (NFLE), median–mean line discriminant analysis (MMLDA), and some other feature extraction approaches in terms of classification accuracy using a small training set.     

Robust input–output linearisation using uncertainty and disturbance estimation

This article proposes a new approach to robustify an input–output linearisation controller. The robustification is achieved by estimating the uncertainties and external unmeasurable disturbances using a novel uncertainty and disturbance estimator. A significant feature of the proposed approach is that it does not need any information about the uncertainties. The stability of the system and the estimator is established. Effectiveness of the proposed approach is demonstrated through application to the wing rock motion control problem.     

A novel minorization–maximization framework for simultaneous feature selection and clustering of high-dimensional count data

Pattern Analysis and Applications - Count data are commonly exploited in machine learning and computer vision applications; however, they often suffer from the well-known curse of dimensionality,...     

An unsupervised method of classifying remotely sensed images using Kohonen self‐organizing maps and agglomerative hierarchical clustering methods

Unlike conventional unsupervised classification methods, such as K‐means and ISODATA, which are based on partitional clustering techniques, the methodology proposed in this work attempts to take advantage of the properties of Kohonen's self‐organizing map (SOM) together with agglomerative hierarchical clustering methods to perform the automatic classification of remotely sensed images. The key point of the proposed method is to execute the cluster analysis process by means of a set of SOM prototypes, instead of working directly with the original patterns of the image. This strategy significantly reduces the complexity of the data analysis, making it possible to use techniques that have not normally been considered viable in the processing of remotely sensed images, such as hierarchical clustering methods and cluster validation indices. Through the use of the SOM, the proposed method maps the original patterns of the image to a two‐dimensional neural grid, attempting to preserve the probability distribution and topology of the input space. Afterwards, an agglomerative hierarchical clustering method with restricted connectivity is applied to the trained neural grid, generating a simplified dendrogram for the image data. Utilizing SOM statistic properties, the method employs modified versions of cluster validation indices to automatically determine the ideal number of clusters for the image. The experimental results show examples of the application of the proposed methodology and compare its performance to the K‐means algorithm.     

Planning and scheduling a fleet of rigs using simulation–optimization

Some of the most important and expensive activities in the oil field development and production phases relate to using rigs. These can be used for drilling wells, or for maintenance activities. As rigs are usually scarce compared to the number of wells requiring service, a schedule of wells to be drilled or repaired must be devised. The objective is to minimize opportunity costs within certain operating constraints. This paper present the first stochastic approach to deals with the problem of planning and scheduling a fleet of offshore oil rigs, where the service time is assumed being uncertain. A simulation–optimization method is used to generate “expected solutions” and performance measures for rigs, as well as statistics about well allocation to rigs. The methodology can be used in two different ways – to schedule an existing fleet of rigs or to scale the size of the fleet – both contemplating the uncertain nature of the problem. The method’s expected results include performance measures for each rig, expected delay for a well to be served, the expected schedule of rigs, and a distribution of the well servicing order. The experiments based on real situations demonstrate the effectiveness of the simulation–optimization approach.     

Optimized gray-scale image watermarking using DWT–SVD and Firefly Algorithm

This paper presents an optimized watermarking scheme based on the discrete wavelet transform (DWT) and singular value decomposition (SVD). The singular values of a binary watermark are embedded in singular values of the LL3 sub-band coefficients of the host image by making use of multiple scaling factors (MSFs). The MSFs are optimized using a newly proposed Firefly Algorithm having an objective function which is a linear combination of imperceptibility and robustness. The PSNR values indicate that the visual quality of the signed and attacked images is good. The embedding algorithm is robust against common image processing operations. It is concluded that the embedding and extraction of the proposed algorithm is well optimized, robust and show an improvement over other similar reported methods.