Parallel Support Vector Machine Research Articles

Identification of facial expression using a multiple impression feedback recognition model

Facial expression recognition (FER) plays a vital role in the automatic detection of human emotions with intelligent machine. Since the FER is an interdisciplinary technique involving biology, computer science and even psychology, more challenges will be encountered as we pursue a high recognition accuracy of facial expressions. Inspired by the progressive enhancing procedure of face recognition of human, we proposed a multiple impression feedback recognition model (MIFR) for the identification of facial expression. Different from current deep learning techniques, the MIFR realizes a quick facial expression recognition through cascade feedback recognition cycles. Multiple impression features of an FER image are firstly obtained by the discrete wavelet decomposition (DWT). Each recognition cycle is implemented by inputting wavelet impression features in a specific decomposition scale into a classifier set of parallel Support Vector Machines (SVMs). In terms of coarse-to-fine wavelet features on multiple scales, the recognition results are gradually improved through integrating classification probability vectors of multiple cycles. In order to validate the performance of the MIFR_SVM for face expression recognition, we also conducted an experiment of facial multi-view expression with occlusion (FMEO) in the laboratory. Three traditional schemes and seven deep learning schemes have been chosen for the FER comparison of three public datasets and one lab dataset FMEO. The classification results show that the MIFR_SVM is not only superior than traditional schemes, but also performs better than deep learning techniques, including both neural-network-based ones and random-forest-based ones. The average recognition accuracy of MIFR_SVM reaches to 92.31% for four datasets. Furthermore, the MIFR_SVM has fewer parameters, less complexity and adaptable cascade structure, which is more suitable for the image datasets with small size or with diverse qualities.

New Sequential and Parallel Support Vector Machine with Grey Wolf Optimizer for Breast Cancer Diagnosis

Breast cancer is one of the most common types of cancer worldwide. Early detection of cancer increases the probability of recovery. This work has three contributions. The first contribution is improving the performance of support vector machine (SVM) using a recent grey wolf optimizer (GWO) for diagnosis breast cancer with efficient scaling techniques. The second contribution is proposing three efficient scaling techniques against the classical normalization technique. The last contribution is using a parallel technique which applies task distribution to improve the efficiency of GWO. The proposed sequential model is applied on two different datasets, Wisconsin diagnosis breast cancer (WDBC) dataset and Electronic Health Records (EHR). Experimental results of WDBC show that the proposed hybrid GWO-SVM model achieves 98.60% with normalization scaling. Also, using the proposed scaling techniques with the proposed GWO-SVM model gives a fast convergence and achieves accuracy rate by 99.30%. The parallel version of the proposed model achieves a speedup by 3.9 on four CPU cores. On the other hand, Experimental results of EHR show that the proposed hybrid GWO-SVM model achieves 93.26% with normalization scaling against 82.05 for SVM.

Neuromechanical Signal-Based Parallel and Scalable Model for Lower Limb Movement Recognition

Individual who have lost their lower limb because of amputation can use the prosthesis to restore daily living activities. The amputee intent recognition during locomotion modes can be used as source to control lower limb prosthesis. Due to continuous data recording from multiple sensors, the timely recognition of activities of daily living have become a challenging issue for traditional technology and conventional machine learning algorithms. This work hypothesize that parallel discriminant features can be learned from large amount of data generated by aggregating the neuromechanical signals from multiple subjects with parallel and distributed computing platform. Consequently, this paper apply three classifiers including support vector machine, decision tree and random forest on large data sets. The model performance is extensively evaluated in terms of different performance measurement parameters such as accuracy, efficiency, scalability and speedup in sequential and distributed environment. The experimental results show that the parallel approach achieved 3.9x computation speedup as compared to the sequential approach without affecting accuracy level. The parallel support vector machine algorithm demonstrated high speedup and scalability in comparison with random forest and decision tree algorithms. The outcome of this study could promote parallel based model for the unobtrusive recognition of lower limb locomotion modes and could promote the future design for the intelligent control of prostheses and exoskeleton.

Research on SVM Remote Sensing Image Classification Based on Parallelization

With the development of technology, the feature dimension and data volume of remote sensing image classification have increased rapidly. However, when remote sensing image classification based on support vector machine (SVM) is used for large-scale data calculation, there are significant limitations in training time. This paper focuses on the parallel processing method of support vector machine (SVM). Based on the popular hybrid parallel support vector machine, a hybrid parallel support vector machine based on sample cross combination is proposed, and carried on the simulation experiment analysis in the stand-alone environment.

Influencing Factors of e‐Commerce Enterprise Development Based on Mobile Computing Big Data Analysis

With the development of increasingly advanced information technology and electronic technology, especially with regard to physical information systems, cloud computing systems, and social services, big data will be widely visible, creating benefits for people and at the same time facing huge challenges. In addition, with the advent of the era of big data, the scale of data sets is getting larger and larger. Traditional data analysis methods can no longer solve the problem of large‐scale data sets, and the hidden information behind big data is digging out, especially in the field of e‐commerce. We have become a key factor in competition among enterprises. We use a support vector machine method based on parallel computing to analyze the data. First, the training samples are divided into several working subsets through the SOM self‐organizing neural network classification method. Compared with the ever‐increasing progress of information technology and electronic equipment, especially the related physical information system finally merges the training results of each working set, so as to quickly deal with the problem of massive data prediction and analysis. This paper proposes that big data has the flexibility of expansion and quality assessment system, so it is meaningful to replace the double‐sidedness of quality assessment with big data. Finally, considering the excellent performance of parallel support vector machines in data mining and analysis, we apply this method to the big data analysis of e‐commerce. The research results show that parallel support vector machines can solve the problem of processing large‐scale data sets. The emergence of data dirty problems has increased the effective rate by at least 70%.

A Robust Multimodal Biometric System VIA Multiple Svms

Generally single Support Vector Machine (SVM) is employed in existing multimodal biometric authentication techniques, and it assumes that whole set of the classifiers is available. But sometimes it is not possible due to some circumstances e.g. injury, some medical treatment etc. This paper includes a robust multimodal biometric authentication system that integrates FKP (Finger-Knuckle Print), face and fingerprint at matching score level fusion using multiple parallel Support Vector Machines (SVMs). Multiple SVMs are applied to overcome the problem of missing biometric modality. Every possible combination of three modalities (FKP, face and fingerprint) are taken into consideration and all combinations have a corresponding SVM to fuse the matching scores and produce the final score set for decision making. Proposed system is more flexible and robust as compared to existing multimodal biometric system with single SVM. The average accuracy of proposed system is estimated on a publicly available dataset with the use of MUBI tool(Multimodal Biometrics Integration tool) and MATLAB 2017b.

Image steganalysis based on convolutional neural network and feature selection

SummarySteganalysis is to detect whether or not the seemly innocent image hiding message. It is an important research topic in information security. With the development of steganography technology, steganalysis becomes more and more difficult. Some steganalysis methods have been proposed to improve the performance. Most research work concentrates on special steganography information detection and the image steganography features are designed manually. Few research works concentrate on universal steganalysis methods. In this paper, as the first several attempts, a novel image steganalysis method based on deep neural network is proposed. First, image high‐frequency features are extracted with wavelet transformation method because that most image hiding message are high frequency. Second, high‐dimensional image steganography features are extracted with deep neural networks according to the high‐frequency images and informative features combination is selected with a novel feature selection method based on entropy. Then, a parallel SVM model is proposed to build the steganalysis model based on large scale training samples. At last, the efficiency of the proposed method is illustrated through analyzing a practical image steganalysis example.

Parallel Implementation on FPGA of Support Vector Machines Using Stochastic Gradient Descent

Sequential Minimal Optimization (SMO) is the traditional training algorithm for Support Vector Machines (SVMs). However, SMO does not scale well with the size of the training set. For that reason, Stochastic Gradient Descent (SGD) algorithms, which have better scalability, are a better option for massive data mining applications. Furthermore, even with the use of SGD, training times can become extremely large depending on the data set. For this reason, accelerators such as Field-programmable Gate Arrays (FPGAs) are used. This work describes an implementation in hardware, using FPGA, of a fully parallel SVM using Stochastic Gradient Descent. The proposed FPGA implementation of an SVM with SGD presents speedups of more than 10,000× relative to software implementations running on a quad-core processor and up to 319× compared to state-of-the-art FPGA implementations while requiring fewer hardware resources. The results show that the proposed architecture is a viable solution for highly demanding problems such as those present in big data analysis.

Particle Swarm Optimization (PSO) with fuzzy c means (PSO‐FCM)–based segmentation and machine learning classifier for leaf diseases prediction

SummaryThis paper proposes an automatic classification technique that uses leaf images some medicinal plants. It is primarily the core reason that drives the research presented here, including the introduction of new innovative segmentation and classification techniques that are deployed to facilitate automatic detection. The major aim of the work is to introduce a new leaf disease prediction technique. The study conducted here a unique but effective image segmentation, feature extraction, as well as plant leaf disease classification. The proposed approach initially preprocesses leaf images of plants thereafter which the diseased sections of the plant are segmented by deploying Particle Swarm Optimization (PSO)–based fuzzy c means segmentation (PSO‐FCM), Gaussian Mixture Model (GMM)–based background subtraction. Vein and shape features, edge‐based feature extraction, and texture characteristics or texture features (TF) are computed. This methodology classifies the leaves of medicinal plants by deploying the Multiple Kernel Parallel Support Vector Machine (MK‐PSVM) classifier. The classifier is implemented via the use of MATLAB classifier. The results are measured using the accuracy, sensitivity, specificity, precision, and F‐measure metrics. Experimental results depict that the classifiers that have been proposed here achieve a higher classification accuracy enabling leaf detection.

A Review of Parallel Support Vector Machines (PSVMs) for Big Data classification

A Review of Parallel Support Vector Machines (PSVMs) for Big Data classification

Analysis of Parallel SVM Based Classification Technique on Healthcare using Big Data Management in Cloud Storage

Analysis of Parallel SVM Based Classification Technique on Healthcare using Big Data Management in Cloud Storage

Dynamic Prediction Research of Silicon Content in Hot Metal Driven by Big Data in Blast Furnace Smelting Process under Hadoop Cloud Platform

In order to explore a dynamic prediction model with good generalization performance of the content of [Si] in molten iron, an improved SVM algorithm is proposed to enhance its practicability in the big data sample set of the smelting process. Firstly, we propose a parallelization scheme to design an SVM solution algorithm based on the MapReduce model under a Hadoop platform to improve the solution speed of the SVM on big data sample sets. Secondly, based on the characteristics of stochastic subgradient projection, the execution time of the SVM solver algorithm does not depend on the size of the sample set, and a structured SVM algorithm based on the neighbor propagation algorithm is proposed, and on this basis, a parallel algorithm for solving the covariance matrix of the training set and a parallel algorithm of the tth iteration of the random subgradient projection are designed. Finally, the historical production big data of No. 1 blast furnace in Tangshan Iron Works II was analyzed during 2015.12.01~2016.11.30 using the reaction mechanism, control mechanism, and gray correlation model in the process of blast furnace iron-making, an essential sample set with input x1k,x2k−3,x3k−3,…,x18k,x19k−1 and output Sik+1 is constructed, and the dynamic prediction model of the content of [Si] in molten iron and the dynamic prediction model of [Si] fluctuation in the molten iron are obtained on the Hadoop platform by means of the structure and parallelized SVM solving algorithm. The results of the research show that the structural and parallel SVM algorithms in the hot metal [Si] content value dynamic prediction hit rate and lifting dynamic prediction hit rate were 91.2% and 92.2%, respectively. Two kinds of dynamic prediction algorithms based on structure and parallelization are 54 times and 5 times faster than traditional serial solving algorithms.

Recognition on Images from Internet Street View Based on Hierarchical Features Learning with CNNs

This article describes hierarchical features with unsupervised learning on images from internet street view images. This is due to the time spent by trained researchers on feature construction steps with traditional methods. This article focuses on the activation of each layer of with convolutional neural networks (CNNs) on Internet street view images detection and compared similarities and differences among them on each layer. The experiment results achieved error rates of 21% on recognition which work went relatively well than the traditional machine learning techniques, such as Parallel SVM.

Timely Identification of Disease by Parallel Real-time Automated Processing of Huge Medical Databases of Images Distributed Geographically, through Knowledge Sharing

Background: The diagnosis of diseases correctly became a challenge, and any error can cost patients life, especially when there is a lack of knowledge or expertise related to a disease, it often results in patient’s death or takes the form of an epidemic, as we have seen in the case of Ebola. Objective: The automation and the development of reliable diagnostic systems became a necessity. Through the use of technology, we can automatically share the knowledge without formal interaction as well as we can identify areas where the disease is spreading while it is not known by the doctors there. Methods: We have presented a complete system that utilizes a combination of one of the best techniques in the field of parallelism, classification, and knowledge sharing. We have used two data sets (DDSM and Belarus Tuberculosis data) to test the applicability of the idea. After retrieving the data, the images are preprocessed, and then Gray level co-occurrence matrix features have been extracted and finally passed to training using three versions of support vector machines. Results: GPU-Accelerated SVM outperformed both parallelized SVM and sequential SVM using breast cancer data, but with lung CT images, GPU-accelerated LIBSVM have not given a remarkable speed-up because the data is small and the gain is lost due to the gpu-cpu memory and cpu-gpu transfer time. The accuracy performances given by three SVMs were identical. Conclusion: Automation through knowledge sharing and parallel computing can help to deal across the world with diseases and it will be easy for doctors to draw the inference. Keywords: Diseases identification, image classification, image processing, knowledge sharing, parallel support vector machines, parallelism.

High-performance Chinese multiclass traffic sign detection via coarse-to-fine cascade and parallel support vector machine detectors

The high variability of sign colors and shapes in uncontrolled environments has made the detection of traffic signs a challenging problem in computer vision. We propose a traffic sign detection (TSD) method based on coarse-to-fine cascade and parallel support vector machine (SVM) detectors to detect Chinese warning and danger traffic signs. First, a region of interest (ROI) extraction method is proposed to extract ROIs using color contrast features in local regions. The ROI extraction can reduce scanning regions and save detection time. For multiclass TSD, we propose a structure that combines a coarse-to-fine cascaded tree with a parallel structure of histogram of oriented gradients (HOG) + SVM detectors. The cascaded tree is designed to detect different types of traffic signs in a coarse-to-fine process. The parallel HOG + SVM detectors are designed to do fine detection of different types of traffic signs. The experiments demonstrate the proposed TSD method can rapidly detect multiclass traffic signs with different colors and shapes in high accuracy.

Adjustable general line coordinates for visual knowledge discovery in n-D data

Preserving all multidimensional data in two-dimensional visualization is a long-standing problem in Visual Analytics, Machine Learning/Data Mining, and Multiobjective Pareto Optimization. While Parallel and Radial (Star) coordinates preserve all n-D data in two dimensions, they are not sufficient to address visualization challenges of all possible datasets such as occlusion. More such methods are needed. Recently, the concepts of lossless General Line Coordinates that generalize Parallel, Radial, Cartesian, and other coordinates were proposed with initial exploration and application of several subclasses of General Line Coordinates such as Collocated Paired Coordinates and Star Collocated Paired Coordinates. This article explores and enhances benefits of General Line Coordinates. It shows the ways to increase expressiveness of General Line Coordinates including decreasing occlusion and simplifying visual pattern while preserving all n-D data in two dimensions by adjusting General Line Coordinates for given n-D datasets. The adjustments include relocating, rescaling, and other transformations of General Line Coordinates. One of the major sources of benefits of General Line Coordinates relative to Parallel Coordinates is twice less number of point and lines in visual representation of each n-D points. This article demonstrates the benefits of different General Line Coordinates for real data visual analysis such as health monitoring and benchmark Iris data classification compared with results from Parallel Coordinates, Radvis, and Support Vector Machine. The experimental part of the article presents the results of the experiment with about 70 participants on efficiency of visual pattern discovery using Star Collocated Paired Coordinates, Parallel, and Radial Coordinates. It shows advantages of visual discovery of n-D patterns using General Line Coordinates subclass Star Collocated Paired Coordinates with n = 160 dimensions.

Latent‐lSVM classification of very high‐dimensional and large‐scale multi‐class datasets

SummaryWe propose a new parallel learning algorithm of latent local support vector machines (SVM), called latent‐lSVM for effectively classifying very high‐dimensional and large‐scale multi‐class datasets. The common framework of texts/images classification tasks using the Bag‐Of‐(visual)‐Words model for the data representation leads to hard classification problem with thousands of dimensions and hundreds of classes. Our latent‐lSVM algorithm performs these complex tasks into two main steps. The first one is to use latent Dirichlet allocation for assigning the datapoint (text/image) to some topics (clusters) with the corresponding probabilities. This aims at reducing the number of classes and the number of datapoints in the cluster compared to the full dataset, followed by the second one: to learn in a parallel way nonlinear SVM models to classify data clusters locally. The numerical test results on nine real datasets show that the latent‐lSVM algorithm achieves very high accuracy compared to state‐of‐the‐art algorithms. An example of its effectiveness is given with an accuracy of 70.14% obtained in the classification of Book dataset having 100 000 individuals in 89 821 dimensional input space and 661 classes in 11.2 minutes using a PC Intel(R) Core i7‐4790 CPU, 3.6 GHz, 4 cores.

Text Detection Model Based on Parallel SVM Algorithm

With the development of the computer science and technology, the advanced algorithm rapidly develops in the field of the text detection and recognition and reduction. However, how to establish a model to be faster and more accurate access to more complex variants or the texts with missing information is to be solved urgently. In this paper, MATLAB parallel computing toolbox is applied to develop the parallel SVM.And this paper improves the operational efficiency through the distributed algorithm and selects the appropriate algorithm parameters to improve the accuracy of the model detection through cross validation. Based on the parallel MATLAB platform of multiple computers, this paper carries out a simulation test and concludes that the accuracy of the complex text detection is 98%. This paper establishes a text detection model that has a high accuracy and a high robustness even in a complex background.

Design and Implementation of a Communication-Optimal Classifier for Distributed Kernel Support Vector Machines

We consider the problem of how to design and implement communication-efficient versions of parallel kernel support vector machines, a widely used classifier in statistical machine learning, for distributed memory clusters and supercomputers. The main computational bottleneck is the training phase, in which a statistical model is built from an input data set. Prior to our study, the parallel isoefficiency of a state-of-the-art implementation scaled as $W=\Omega (P^3)$ , where $W$ is the problem size and $P$ the number of processors; this scaling is worse than even a one-dimensional block row dense matrix vector multiplication, which has $W=\Omega (P^2)$ . This study considers a series of algorithmic refinements, leading ultimately to a Communication-Avoiding SVM method that improves the isoefficiency to nearly $W=\Omega (P)$ . We evaluate these methods on 96 to 1,536 processors, and show average speedups of $3-16\times$ ( $7\times$ on average) over Dis-SMO, and a 95 percent weak-scaling efficiency on six real-world datasets, with only modest losses in overall classification accuracy. The source code can be downloaded at [1] .

A combined parallel genetic algorithm and support vector machine model for breast cancer detection

A combined parallel genetic algorithm and support vector machine model for breast cancer detection